CN119666666B - Automatic detection method and system applied to freshness of fruits and vegetables - Google Patents
Automatic detection method and system applied to freshness of fruits and vegetablesInfo
- Publication number
- CN119666666B CN119666666B CN202510152196.4A CN202510152196A CN119666666B CN 119666666 B CN119666666 B CN 119666666B CN 202510152196 A CN202510152196 A CN 202510152196A CN 119666666 B CN119666666 B CN 119666666B
- Authority
- CN
- China
- Prior art keywords
- detected
- fruit
- vegetable
- edge
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention relates to the technical field of fruit and vegetable freshness detection, in particular to an automatic detection method and system applied to fruit and vegetable freshness, and the method comprises the steps of obtaining a gray level image of fruits and vegetables to be detected, obtaining edge clusters of the fruits and vegetables to be detected according to the gray level image, calculating the covering degree of the edge clusters of the fruits and vegetables to be detected, obtaining particle density of the edge clusters of the fruits and vegetables to be detected according to the covering degree, further obtaining plumpness of the fruits and vegetables to be detected according to the particle density, obtaining pixel points of the fruits and vegetables to be detected in the edge clusters of the fruits and vegetables to be detected, selecting and calculating a loss value of the pixel points of the fruits and vegetables to be detected, calculating sugar analysis degree of the fruits and vegetables to be detected according to the pixel points of the fruits and vegetables to be detected, and obtaining freshness of the fruits and vegetables to be detected according to the plumpness and sugar analysis degree, wherein the freshness of the fruits and vegetables to be detected is not qualified when the freshness is lower than a preset fresh threshold. By adopting the method, the freshness of the fruits and vegetables can be detected more accurately.
Description
Technical Field
The invention relates to the technical field of fruit and vegetable freshness detection, in particular to an automatic detection method and system applied to fruit and vegetable freshness.
Background
In the industrial field, fruits and vegetables are often concentrated and processed into jam, and the process can not only remarkably improve the flavor of the jam, but also make the jam a multifunctional food ingredient. The production of jam is strictly dependent on the freshness of the fruit and vegetable raw materials, which refers to the capability of the fruit and vegetable to maintain the best taste, nutritional value and appearance of the fruit and vegetable in the steps from picking, storing and transporting. In view of the fact that freshness is directly related to the quality of fruits and vegetables and the eating experience of consumers, it is particularly important to detect freshness of raw materials before jam production.
In order to ensure the quality of jam, taking grape as an example, the freshness detection of the grape becomes a key step. By deploying a high-resolution camera on the grape conveyor belt, images of the grapes can be automatically acquired for detection. However, existing detection methods perform well in the face of significant changes in grape color, but remain challenging when identifying more concealed signs of freshness such as moisture loss or slight sugar separation.
Disclosure of Invention
In order to solve the technical problem that the freshness detection of fruits and vegetables is not accurate enough in the prior art, the invention aims to provide an automatic detection method for the freshness of fruits and vegetables, and the adopted technical scheme is as follows:
Acquiring a gray level image of fruits and vegetables to be detected, and acquiring edge clusters of the fruits and vegetables to be detected according to the gray level image;
calculating the covering degree of the fruit and vegetable grain edge cluster to be detected, and acquiring the grain density of the fruit and vegetable grain edge cluster to be detected according to the covering degree, so as to acquire the plumpness of the fruit and vegetable to be detected according to the grain density;
acquiring to-be-detected fruit and vegetable particle pixel points in the to-be-detected fruit and vegetable particle edge clusters, selecting and calculating a loss value of the to-be-detected fruit and vegetable particle pixel points, and calculating sugar analysis degree of the to-be-detected fruit and vegetable according to the to-be-detected fruit and vegetable particle pixel points corresponding to the smallest loss value;
And acquiring the freshness of the fruits and vegetables to be detected according to the plumpness and the sugar analysis degree, wherein when the freshness is lower than a preset freshness threshold value, the freshness of the fruits and vegetables to be detected is unqualified.
Further, obtaining the fruit and vegetable grain edge cluster to be detected according to the gray level image comprises the following steps:
Performing edge detection on the gray level image to obtain an edge image, and obtaining a fruit and vegetable connected domain to be detected according to the edge image;
Performing circle fitting on the edges of fruit and vegetable particles to be detected in the fruit and vegetable communication domain to be detected by adopting a least square method to obtain a fitted circle, and acquiring a fitting circle center and a fitting radius of the fitted circle;
Searching for a target fruit and vegetable particle edge to be detected, the distance between the target fruit and vegetable particle edge and the corresponding fitting circle center of which is smaller than a preset first distance value, and screening out the fruit and vegetable particle edge to be detected, corresponding to the fitting radius, of which the absolute value of the difference value between the target fruit and vegetable particle edge to be detected and the fitting radius of which is larger than a preset second distance value, from the target fruit and vegetable particle edge to be detected, so as to obtain a co-circle edge, wherein the fitting radius corresponding to the target fruit and vegetable particle edge to be detected is the target fitting radius;
Calculating a first circle fitting loss value of the fruit and vegetable grain edges to be detected and a second circle fitting loss value which is common to the fruit and vegetable grain edges to be detected and the corresponding co-circular edges, acquiring an edge coefficient according to the first circle fitting loss value and the second circle fitting loss value, and when the edge coefficient is larger than a preset edge threshold value, the fruit and vegetable grain edges to be detected are fruit and vegetable grain edge clusters.
Further, obtaining an edge coefficient from the first circle fit loss value and the second circle fit loss value includes:
The inverse of the first circle fit loss value multiplied by the value of the second circle fit loss value is used as an input value of an exponential function based on e, and the output value of the exponential function is used as the edge coefficient.
Further, the obtaining the coverage of the fruit and vegetable grain edge cluster to be detected comprises the following steps:
Acquiring the fitting radius and the edge length sum of the edge clusters of the fruit and vegetable particles to be detected;
And obtaining a fitting circle circumference according to the fitting radius, wherein the difference value of the sum of the edge lengths subtracted from the fitting circle circumference is used as an input value of a hyperbolic function, and an output value of the hyperbolic function is used as the covering degree of the edge cluster of the fruit and vegetable to be detected.
Further, obtaining the particle density of the fruit and vegetable particle edge clusters to be detected according to the coverage comprises the following steps:
Multiplying the area of the fitting circle corresponding to the fruit and vegetable particles to be detected by the covering degree corresponding to the fruit and vegetable particles to be detected to obtain a first product;
and repeating the first product obtaining process to obtain the first products corresponding to the fruit and vegetable grains to be detected, and subtracting the area of the fruit and vegetable connected domain to be detected after sequentially adding the first products to obtain the grain density of the edge clusters of the fruit and vegetable grains to be detected.
Further, obtaining the plumpness of the fruits and vegetables to be detected according to the particle density comprises:
obtaining a loss value of a fitting circle of the fruit and vegetable grain edge cluster to be detected and a minimum distance between the fruit and vegetable grain edge cluster to be detected and a surrounding edge cluster, wherein the value obtained by dividing the loss value by the minimum distance is used as a first ratio;
repeating the first ratio obtaining process to obtain the first ratio corresponding to each fruit and vegetable grain edge cluster to be detected, and obtaining an average value of the first ratio;
And taking the average value as an input value of a hyperbolic function, and multiplying the output value of the hyperbolic function by the particle density to obtain the plumpness of the fruits and vegetables to be detected.
Further, selecting and calculating the loss value of the fruit and vegetable particle pixel points to be detected comprises the steps of obtaining the gray average value of the fruit and vegetable particle pixel points to be detected in the fruit and vegetable particle edge cluster to be detected, obtaining the difference value between the gray value of each fruit and vegetable particle pixel point to be detected and the gray average value, and sequencing the fruit and vegetable particle pixel points to be detected according to the difference value from large to small;
selecting any number of the fruit and vegetable particle pixels to be detected with the front sequence as white point pixels, and selecting any number of the fruit and vegetable particle pixels to be detected with the front sequence as light spot pixels;
the vector between the spot pixel point and the circle center of the fitting circle corresponding to the spot pixel point is recorded as a spot vector;
Dividing the variance of the gray value of the white point pixel point by the information entropy of the light spot vector to obtain a second ratio, multiplying the opposite number of the second ratio by the gray average value of the white point pixel point to be used as an input value of an exponential function based on e, and using the output value of the exponential function as a loss value of the fruit and vegetable particle pixel point to be detected.
Further, calculating the sugar analysis degree of the fruits and vegetables to be detected according to the pixels of the fruits and vegetables to be detected corresponding to the minimum loss value comprises:
The white point pixel point corresponding to the smallest loss value is marked as a white frost pixel point, and a distance average value between the white frost pixel point and the edge of the fruit and vegetable to be detected and a gradient value of the white frost pixel point are obtained;
calculating the absolute value of the gradient value of the adjacent white frost pixel point subtracted by the gradient value of the white frost pixel point in the adjacent white frost pixel point of the white frost pixel point, wherein the adjacent white frost pixel point corresponding to the minimum absolute value is connected with the white frost pixel point to obtain a gradient line;
Multiplying the length of the gradient line by the gradient value of the gradient line, and dividing the length by the loss value of the gradient line obtained by fitting a straight line by using a least square method to obtain a second ratio;
Repeating the second ratio obtaining process to obtain the second ratio corresponding to each gradient line, and obtaining an average value of the second ratio;
And multiplying the reciprocal of the distance average value by the average value of the second ratio to obtain an input value of a hyperbolic function, wherein the output value of the hyperbolic function is used as the sugar analysis degree of the fruits and vegetables to be detected.
Further, obtaining the freshness of the fruit and vegetable to be detected according to the plumpness and the sugar analysis degree comprises:
Dividing the fullness by the sugar analysis degree to obtain the freshness of the fruits and vegetables to be detected.
The embodiment of the invention also provides an automatic detection system applied to the freshness of fruits and vegetables, and the method comprises the following steps:
the acquisition module is used for acquiring a gray level image of fruits and vegetables to be detected, and acquiring edge clusters of the fruits and vegetables to be detected according to the gray level image;
The plumpness module is used for calculating the covering degree of the fruit and vegetable grain edge cluster to be detected, acquiring the grain density of the fruit and vegetable grain edge cluster to be detected according to the covering degree, and further acquiring the plumpness of the fruit and vegetable to be detected according to the grain density;
The sugar analysis degree module is used for acquiring the fruit and vegetable particle pixel points to be detected in the fruit and vegetable particle edge cluster to be detected, selecting and calculating the loss value of the fruit and vegetable particle pixel points to be detected, and calculating the sugar analysis degree of the fruit and vegetable to be detected according to the fruit and vegetable particle pixel points to be detected corresponding to the smallest loss value;
and the freshness module is used for acquiring the freshness of the fruits and vegetables to be detected according to the plumpness and the sugar analysis degree, and when the freshness is lower than a preset freshness threshold value, the freshness of the fruits and vegetables to be detected is unqualified.
The invention has the following beneficial effects:
Firstly, a gray level image of fruits and vegetables to be detected is obtained, and edge clusters of the fruits and vegetables to be detected are obtained according to the gray level image. Each fruit and vegetable grain edge cluster to be detected represents one fruit and vegetable grain. Secondly, calculating the covering degree of the fruit and vegetable grain edge clusters to be detected, acquiring the grain density of the fruit and vegetable grain edge clusters to be detected according to the covering degree, and further acquiring the plumpness of the fruit and vegetable to be detected according to the grain density. The fullness is a key index for detecting the freshness of fruits and vegetables, and the freshness of fruits and vegetables is higher.
And then, obtaining the pixel points of the fruit and vegetable particles to be detected in the edge cluster of the fruit and vegetable particles to be detected, selecting and calculating the loss value of the pixel points of the fruit and vegetable particles to be detected, and calculating the sugar analysis degree of the fruit and vegetable particles to be detected according to the pixel points of the fruit and vegetable particles to be detected corresponding to the minimum loss value. The sugar analysis degree is also a key index for detecting the freshness of fruits and vegetables, and the more fresh the fruits and vegetables are, the lower the sugar analysis degree is. And finally, acquiring the freshness of the fruits and vegetables to be detected according to the plumpness and the sugar analysis degree, wherein when the freshness is lower than a preset freshness threshold value, the freshness of the fruits and vegetables to be detected is unqualified. By adopting the method, when slight unreliability of fruits and vegetables such as water loss or sugar analysis occurs, the unqualified freshness of the fruits and vegetables can still be detected, so that the freshness detection of the fruits and vegetables is more accurate.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of an automatic detection method for freshness of fruits and vegetables according to a first embodiment of the present invention;
Fig. 2 is a flowchart of acquiring a cluster of edges of fruits and vegetables to be detected according to the gray level chart according to a second embodiment of the present invention;
Fig. 3 is a flowchart for obtaining the coverage of the edge cluster of the fruit and vegetable grain to be detected according to a third embodiment of the present invention;
fig. 4 is a flowchart of obtaining a particle density of the edge clusters of the fruit and vegetable particles to be detected according to the coverage according to a fourth embodiment of the present invention;
Fig. 5 is a flowchart of obtaining the plumpness of the fruits and vegetables to be detected according to the particle density according to a fifth embodiment of the present invention;
FIG. 6 is a flowchart of selecting and calculating a loss value of pixels of the fruit and vegetable particles to be detected according to a sixth embodiment of the present invention;
Fig. 7 is a flowchart of calculating a sugar analysis degree of the fruits and vegetables to be detected according to the pixels of the fruits and vegetables to be detected corresponding to the smallest loss value according to the seventh embodiment of the present invention;
fig. 8 is a schematic diagram of an automatic detection system for freshness of fruits and vegetables according to an eighth embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following description refers to the specific implementation, structure, characteristics and effects of the automatic detection method for the freshness of fruits and vegetables according to the invention, which are provided by the invention, with reference to the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the automatic detection method applied to the freshness of fruits and vegetables provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of an automatic detection method for freshness of fruits and vegetables according to an embodiment of the present invention is shown, where the method includes:
S101, acquiring a gray level image of fruits and vegetables to be detected, and acquiring edge clusters of the fruits and vegetables to be detected according to the gray level image.
The fruits and vegetables to be detected comprise grape, grape-like grape and other similar string-shaped nearly circular fruits and vegetables.
In the process of producing jam, ensuring freshness of fruits and vegetables is one of the core elements for maintaining high quality of jam. The fresh fruits and vegetables not only can endow the jam with better taste, rich nutrition and attractive color, but also can effectively reduce the deterioration and spoilage risks during storage. Therefore, in the early stage of production, it is important to carry out strict freshness detection on purchased fruits and vegetables.
The production flow of the jam usually comprises a plurality of key links such as raw material preparation, processing, boiling and forming, cooling and solidifying, packaging and sealing, storage and preservation and the like. Particularly, in the raw material preparation stage, a freshness detection program is required to be executed for purchased fruit and vegetable raw materials. This step involves unpacking and placing the fruit and vegetable material on a conveyor for subsequent processing. For accurate detection of freshness, an industrial-grade high-resolution camera can be equipped on the conveyor belt, and is specially used for capturing images of cleaned fruits and vegetables.
In order to scientifically evaluate the freshness of fruits and vegetables, the collected images need to be further subjected to gray scale treatment and converted into gray scale images which are easier to analyze so as to accurately detect the freshness.
For the gray level map obtained by the operation, a plurality of fruit and vegetable strings can be contained. Therefore, in order to detect freshness of each fruit and vegetable string, firstly, edge detection such as canny edge detection needs to be performed on the obtained fruit and vegetable string gray level image to obtain edges in the gray level image, and because of large gray level difference between the fruit and vegetable string and the conveyor belt, the fruit and vegetable string connected domain in the gray level image and the edges inside the fruit and vegetable string connected domain can be obtained.
For the fruit and vegetable cluster connected domain obtained by the operation, the plumping condition of fruit and vegetable grains needs to be analyzed for detecting the freshness of the fruits and vegetables. When the fruit and vegetable strings are fresh, the fruit and vegetable grains are plump and gather, and as time increases, the fruit and vegetable branches in the fruit and vegetable strings and the water in the fruit and vegetable grains run off to cause the condition that the fruits and vegetables are not plump, so that plumpness analysis is required according to the fruit and vegetable grains in the fruit and vegetable strings, but the fruit and vegetable grain edges, the fruit and vegetable grain surfaces and the fruit and vegetable branch and leaf edges exist in the fruit and vegetable string communication domain, and the fruit and vegetable grain edges are required to be screened from the fruit and vegetable string communication domain. Because the fruit and vegetable particles in the fruit and vegetable string communication domain are nearly circular, and more than one grape particle exists in the fruit and vegetable string communication domain, the edges of the fruit and vegetable particles are required to be obtained according to the distribution of the edges in the fruit and vegetable string communication domain, and freshness detection is carried out according to the distribution condition of the edges of the fruit and vegetable particles.
For the inner edge of the fruit and vegetable cluster connected domain, the fruit and vegetable particles are approximately round, so that the nearly circularity of each edge can be determined by fitting a circle. However, since the edges of the fruit and vegetable particles in the fruit and vegetable strings are not completely complete, there are a plurality of edges which are the same fruit and vegetable particles, and since the fruit and vegetable particles are approximately round, the analysis of the edges of the fruit and vegetable particles can be performed by analyzing the co-circularity between the edges.
The process of obtaining the edge clusters of the fruits and vegetables to be detected according to the gray level map will be described in detail in the second embodiment, and will not be described herein.
S102, calculating the covering degree of the edge clusters of the fruit and vegetable particles to be detected, acquiring the particle density of the edge clusters of the fruit and vegetable particles to be detected according to the covering degree, and further acquiring the plumpness of the fruit and vegetable particles to be detected according to the particle density.
Because the fruits and vegetables to be detected grow continuously and become full in the growing process, the distance between the fruits and vegetables is reduced after the fruits and vegetables on the fruit and vegetable clusters become full. In the process that fruit and vegetable particles to be detected become larger gradually, the distance between the fruit and vegetable particles to be detected is reduced, so that extrusion conditions among the fruit and vegetable particles to be detected can be caused, slight deformation of the fruit and vegetable particles to be detected can be caused, and some fruit and vegetable particles to be detected can be covered. When the freshness of fruit and vegetable particles to be detected is reduced, the water in the fruit and vegetable particles to be detected and the water in fruit and vegetable branches to be detected are lost, so that the plumpness of the fruit and vegetable particles to be detected is reduced, the extrusion condition among the fruit and vegetable particles to be detected is eliminated, and the supporting effect of the fruit and vegetable particles to be detected is poor due to the water loss of the fruit and vegetable branches to be detected, so that the situation that the fruit and vegetable particles to be detected sink is caused, and therefore, the plumpness analysis of the fruit and vegetable particles to be detected can be performed through the analysis of the edge clusters of the fruit and vegetable particles to be detected, which are obtained through the operation.
The process of calculating the coverage of the edge clusters of the fruit and vegetable particles to be detected will be described in detail in the third embodiment, and will not be described herein.
The process of obtaining the particle density of the fruit and vegetable particle edge cluster to be detected according to the coverage is described in detail in the fourth embodiment, and is not described herein.
The process of obtaining the fullness of the fruits and vegetables to be detected according to the particle density will be described in detail in the fifth embodiment, and will not be described herein.
S103, obtaining pixel points of fruit and vegetable particles to be detected in the edge clusters of the fruit and vegetable particles to be detected, selecting and calculating a loss value of the pixel points of the fruit and vegetable particles to be detected, and calculating the sugar analysis degree of the fruit and vegetable particles to be detected according to the pixel points of the fruit and vegetable particles to be detected, corresponding to the minimum loss value.
When the high temperature or the storage condition is improper, sugar in the fruits and vegetables to be detected can be gradually separated out, so that white sugar crystallization appears on the surfaces of the fruits and vegetables to be detected, and the freshness of the fruits and vegetables to be detected is further affected. It is known that the higher the sugar analysis degree, the less fresh the fruits and vegetables to be detected.
The fresh fruit and vegetable surface to be detected is covered with a layer of fine and powdery white substances, which is a natural waxy protection layer secreted by the fruit and vegetable to be detected, and the waxy protection layer is helpful for reducing water evaporation and keeping the freshness and water content of the fruit and vegetable to be detected. The waxy protective layer on the surface of the fruit and vegetable to be detected is white, and the waxy protective layer is higher in gray value in a gray level chart. When the smooth part of the surface of the fruit and vegetable to be detected reflects light, the light can be expressed as a higher gray value in the gray map. Therefore, in the gray level diagram, the wax protective layer is confused with the reflection of the surface of the fruit and vegetable to be detected, and the wax protective layer and the reflection of the fruit and vegetable need to be distinguished according to the difference between the wax protective layer and the reflection, so that the white frost degree, which refers to the white frost produced by sugar analysis, is obtained.
The wax protective layer is a fine powdery white substance and different in thickness at different positions, so that the wax protective layer on the surface of fruits and vegetables to be detected is rough and is distributed randomly, the gray value of the wax protective layer is inconsistent due to the fact that the thickness of the wax protective layer is different, reflection generated in the region of the surface of the fruits and vegetables to be detected, which lacks the wax protective layer, is influenced by light conditions, the reflection is relatively concentrated on the surface of the fruits and vegetables to be detected, and the formed light spot shapes tend to be consistent.
The method comprises the steps of obtaining to-be-detected fruit and vegetable particle pixel points in the to-be-detected fruit and vegetable particle edge clusters, firstly obtaining to-be-detected fruit and vegetable particle pixel points according to each to-be-detected fruit and vegetable particle edge in the obtained to-be-detected fruit and vegetable particle edge clusters, and obtaining the circle center of the nearest to any pixel point in the to-be-detected fruit and vegetable particle edge cluster, if the pixel point is in the radius range of the circle center and can be surrounded by the edge, marking the pixel point as to-be-detected fruit and vegetable particle pixel points in the to-be-detected fruit and vegetable particle edge clusters, so that a plurality of to-be-detected fruit and vegetable particle pixel points and to-be-detected fruit and vegetable particle edge clusters can be obtained.
The process of selecting and calculating the loss value of the pixel point of the fruit and vegetable particle to be detected will be described in detail in the sixth embodiment, and will not be described herein.
The process of calculating the sugar analysis degree of the fruits and vegetables to be detected according to the pixels of the fruits and vegetables to be detected corresponding to the smallest loss value will be described in detail in the seventh embodiment, and will not be described herein.
S104, acquiring the freshness of the fruits and vegetables to be detected according to the plumpness and the sugar analysis degree, and if the freshness is lower than a preset freshness threshold value, failing to be detected.
Further, obtaining the freshness of the fruit and vegetable to be detected according to the plumpness and the sugar analysis degree comprises:
Dividing the fullness by the sugar analysis degree to obtain the freshness of the fruits and vegetables to be detected.
It is known that the higher the fullness and the lower the sugar analysis, the fresher the fruit and vegetable.
And when the freshness is lower than a preset freshness threshold, the freshness of the fruits and vegetables to be detected is not qualified. And when the freshness is higher than a preset freshness threshold, the freshness of the fruits and vegetables to be detected is qualified.
The freshness threshold may be set autonomously, preferably to 0.7.
The fruit picking machine arranged on the follow-up conveyor belt can separate fruits and vegetables with qualified freshness from fruits and vegetables with unqualified freshness detected, and the fruits and vegetables on the separated conveyor belt are all with qualified freshness and can be used for manufacturing follow-up jams.
Fig. 2 is a flowchart of obtaining a fruit and vegetable edge cluster to be detected according to the gray level diagram according to a second embodiment of the present invention, where obtaining the fruit and vegetable edge cluster to be detected according to the gray level diagram includes:
S201, edge detection is carried out on the gray level image to obtain an edge image, and the fruit and vegetable connected domain to be detected is obtained according to the edge image.
Edge detection is known in the art as canny edge detection.
S202, performing circle fitting on the edges of fruit and vegetable particles to be detected in the fruit and vegetable communication domain to be detected by adopting a least square method to obtain a fitting circle, and acquiring a fitting circle center and a fitting radius of the fitting circle.
S203, searching for a target fruit and vegetable particle edge to be detected, wherein the distance between the target fruit and vegetable particle edge and the corresponding fitting circle center is smaller than a preset first distance value, and screening out the fruit and vegetable particle edge to be detected, corresponding to the fitting radius, of which the absolute value of the difference value between the target fruit and vegetable particle edge and the fitting radius is larger than a preset second distance value, so as to obtain a co-circle edge, and the fitting radius corresponding to the target fruit and vegetable particle edge to be detected is the target fitting radius.
The method is characterized in that the edge clusters of the fruit and vegetable to be detected are screened from the fruit and vegetable by utilizing the co-circularity of the edges of the fruit and vegetable to be detected. Each fruit and vegetable grain to be detected represents a fruit and vegetable grain roughly at the edge.
S204, calculating a first circle fitting loss value of the fruit and vegetable grain edges to be detected and a second circle fitting loss value which is common to the fruit and vegetable grain edges to be detected and the corresponding co-circle edges, acquiring edge coefficients according to the first circle fitting loss value and the second circle fitting loss value, and when the edge coefficients are larger than a preset edge threshold value, the fruit and vegetable grain edges to be detected are fruit and vegetable grain edge clusters.
The edge threshold may be set autonomously, preferably at 0.7.
Obtaining an edge coefficient from the first circle fit loss value and the second circle fit loss value may be expressed as:
;
Wherein the said Represent the firstThe first circle of edges fits the loss value, theRepresent the firstThe second circle of edges fits the loss value, theRepresent the firstThe edge coefficients of the edges, theAn exponential function based on a natural constant e is represented.
In order to obtain the plumpness of fruits and vegetables to be detected, firstly, carrying out circle fitting on the edges in each obtained edge cluster of fruits and vegetables to be detected to obtain the center of the circle marked as the center of the fruits and vegetables to be detected, wherein different fruits and vegetables to be detected correspond to different fruits and vegetables to be detected due to different edge clusters of fruits and vegetables to be detected, and the extrusion and shielding conditions of different fruits and vegetables to be detected by other fruits and vegetables to be detected are different. And the subsequent analysis of plumpness needs to be treated differently. Therefore, the covering degree of each fruit and vegetable grain edge cluster to be detected needs to be obtained according to the completeness degree of the edges in each fruit and vegetable grain edge cluster to be detected.
Fig. 3 is a flowchart of obtaining the coverage of the edge cluster of the fruit and vegetable grain to be detected according to a third embodiment of the present invention, where the obtaining the coverage of the edge cluster of the fruit and vegetable grain to be detected includes:
s301, acquiring the fitting radius and the edge length sum of the edge clusters of the fruit and vegetable particles to be detected.
S302, obtaining a fitting circle circumference according to the fitting radius, wherein the difference value of the sum of the edge lengths subtracted from the fitting circle circumference is used as an input value of a hyperbolic function, and an output value of the hyperbolic function is used as the covering degree of the edge cluster of the fruit and vegetable to be detected.
The covering degree of the fruit and vegetable grain edge cluster to be detected can be expressed as follows:
;
Wherein the said Indicating the covering degree of the edge cluster of the fruit and vegetable to be detected, whereinRepresenting the fitted circumference, theRepresenting the sum of the lengths of the edges,And the subscript N represents the nth fruit and vegetable grain edge cluster to be detected, and the value is 1 to the number N of the fruit and vegetable grain edge clusters to be detected.
Fig. 4 is a flowchart of obtaining a particle density of the edge cluster of the fruit and vegetable particles to be detected according to the coverage according to the fourth embodiment of the present invention, where the obtaining the particle density of the edge cluster of the fruit and vegetable particles to be detected according to the coverage includes:
S401, multiplying the area of the fitting circle corresponding to the fruit and vegetable particles to be detected by the covering degree corresponding to the fruit and vegetable particles to be detected to obtain a first product.
The first product may be expressed as:
;
Wherein the said Representing the area of the fitting circle corresponding to the nth fruit and vegetable particle to be detected, the method comprises the following steps ofAnd (3) representing the covering degree corresponding to the nth fruit and vegetable grain to be detected, wherein the value of N is 1 to the number N of the edge clusters of the fruit and vegetable grain to be detected.
S402, repeating the first product obtaining process to obtain the first products corresponding to the fruit and vegetable particles to be detected, sequentially adding the first products, and subtracting the area of the fruit and vegetable connected domain to be detected to obtain the particle density of the edge clusters of the fruit and vegetable particles to be detected.
The particle density of the fruit and vegetable particle edge clusters to be detected can be expressed as follows:
;
Wherein the said Representing the particle density of the edge clusters of the fruit and vegetable particles to be detected, the fruit and vegetable particlesAnd the area of the fruit and vegetable connected domain to be detected is represented.
Fig. 5 is a flowchart of obtaining the plumpness of the fruits and vegetables to be detected according to the particle density according to the fifth embodiment of the present invention, where the obtaining the plumpness of the fruits and vegetables to be detected according to the particle density includes:
s501, obtaining a loss value of a fitting circle of the fruit and vegetable grain edge cluster to be detected and a minimum distance between the fruit and vegetable grain edge cluster to be detected and a surrounding edge cluster, wherein the value obtained by dividing the loss value by the minimum distance is used as a first ratio.
The first ratio may be expressed as: Wherein the said Representing the loss value of the fitting circle of the fruit and vegetable grain edge cluster to be detected, the saidAnd representing the minimum distance between the edge cluster of the fruit and vegetable particles to be detected and the peripheral edge cluster.
And obtaining the loss value of the fitting circle of the fruit and vegetable particle edge cluster to be detected by adopting a least square method.
It should be noted that, the minimum distance between the edge cluster of the fruit and vegetable particle to be detected and the surrounding edge cluster refers to the minimum distance between the circle center of the fitting circle of the edge cluster of the fruit and vegetable particle to be detected and the circle center of the fitting circle of the surrounding edge cluster.
S502, repeating the first ratio obtaining process to obtain the first ratio corresponding to each fruit and vegetable grain edge cluster to be detected, and obtaining an average value of the first ratio.
The average of the first ratio may be expressed as: Wherein N represents the number of the fruit and vegetable grain edge clusters to be detected.
S503, taking the average value as an input value of a hyperbolic function, and multiplying the output value of the hyperbolic function by the particle density to obtain the plumpness of the fruits and vegetables to be detected.
The fullness of the fruits and vegetables to be detected can be expressed as follows:
;
Wherein the said Representing the fullness of the fruits and vegetables to be detected, the fruits and vegetables to be detectedRepresenting the particle density, theRepresenting a hyperbolic function.
Fig. 6 is a flowchart for selecting and calculating a loss value of a pixel point of a fruit and vegetable particle to be detected according to a sixth embodiment of the present invention, where selecting and calculating a loss value of a pixel point of a fruit and vegetable particle to be detected includes s601, obtaining a gray average value of the pixel point of the fruit and vegetable particle to be detected in an edge cluster of the fruit and vegetable particle to be detected, obtaining a difference value between the gray value of each pixel point of the fruit and vegetable particle to be detected and the gray average value, and sorting the pixel points of the fruit and vegetable particle to be detected according to the difference value from large to small.
The pixel points of the fruit and vegetable particles to be detected corresponding to the large difference value are arranged in front, and the pixel points of the fruit and vegetable particles to be detected corresponding to the small difference value are arranged in back.
S602, selecting any number of pixels of the fruit and vegetable particles to be detected, which are ranked forward, as white point pixels, and selecting any number of pixels of the fruit and vegetable particles to be detected, which are ranked forward, as light spot pixels.
S603, marking a vector between the spot pixel point and the circle center of the corresponding fitting circle as a spot vector.
S604, dividing the variance of the gray value of the white point pixel point by the information entropy of the light spot vector to obtain a second ratio, multiplying the opposite number of the second ratio by the gray average value of the white point pixel point to be used as an input value of an exponential function based on e, and using the output value of the exponential function as the loss value of the fruit and vegetable particle pixel point to be detected.
The loss value of the pixel points of the fruit and vegetable particles to be detected can be expressed as follows:
;
Wherein the said Representing the loss value of the pixel points of the fruit and vegetable particles to be detected, whereinRepresenting a variance of gray values of the white point pixels, theEntropy representing the spot vector, theRepresenting the gray average of the white point pixels,An exponential function based on e is represented.
The ratio between the variance of gray values of the white point pixels and the information entropy of the light spot vectorThe larger the value, the larger the variance of the gray value of the white point pixel point and the smaller the information entropy of the facula vector, and the larger the white point pixel is the rough white particulate matter generated by white frost. The spot vector represents the position parameter of the spot pixel point in the fruit and vegetable grains to be detected, and the smaller the information entropy of the spot vector is, the more similar the position of the spot pixel point in the fruit and vegetable grains to be detected is, and the more the position condition of the spot is met.
The white-spot pixel point corresponding to the smallest loss value is the white-frost pixel point, and for the obtained white-frost pixel point, the white frost on the surface of the fruit and vegetable to be detected can be secreted fruit wax or fructose which is precipitated improperly due to preservation, wherein the secretion of the fruit wax is random and the fruit wax can be randomly distributed on the surface of the fruit and vegetable to be detected, the fructose can be precipitated from the joint of the fruit and vegetable branches to be detected, and the fructose particles are larger and have obvious particles.
Fig. 7 is a flowchart of calculating a sugar analysis degree of the fruit and vegetable to be detected according to the fruit and vegetable pixel to be detected corresponding to the smallest loss value according to the seventh embodiment of the present invention, where calculating the sugar analysis degree of the fruit and vegetable to be detected according to the fruit and vegetable pixel to be detected corresponding to the smallest loss value includes:
S701, marking the white point pixel point corresponding to the smallest loss value as a white frost pixel point, and acquiring a distance average value between the white frost pixel point and the edge of the fruit and vegetable to be detected and a gradient value of the white frost pixel point.
The distance between the white frost pixel point and the edge of the fruit and vegetable grain to be detected refers to the minimum distance between the white frost pixel point and the edge of the fruit and vegetable grain to be detected.
S702, calculating the absolute value of the gradient value of the adjacent white frost pixel point subtracted from the gradient value of the white frost pixel point in the adjacent white frost pixel point of the white frost pixel point, and connecting the adjacent white frost pixel point corresponding to the minimum absolute value with the white frost pixel point to obtain a gradient line.
S703, dividing the length of the gradient line by the gradient value of the gradient line, and obtaining a second ratio by using a loss value obtained by fitting a straight line to the gradient line by using a least square method.
The second ratio may be expressed as: Wherein the said Representing the length of the gradient line, theA gradient value representing the gradient line, theRepresenting the loss value of the gradient line obtained by fitting a straight line using a least squares method.
S704, repeating the second ratio obtaining process to obtain the second ratio corresponding to each gradient line, and obtaining an average value of the second ratio.
The average of the second ratio may be expressed as: Wherein, X represents the number of the gradient lines.
And S705, multiplying the reciprocal of the distance average value by the average value of the second ratio to obtain an input value of a hyperbolic function, wherein the output value of the hyperbolic function is used as the sugar analysis degree of the fruits and vegetables to be detected.
The sugar analysis degree of the fruits and vegetables to be detected can be expressed as follows:
;
Wherein the said Indicating the sugar analysis degree of the fruits and vegetables to be detected, theRepresenting the inverse of the distance mean, theRepresenting a hyperbolic function.
And for the distance between the white frost pixel point and the edge of the fruit and vegetable to be detected, the smaller the value is, the closer the white frost is to the outside of the fruit and vegetable particles to be detected in the image, and the more belongs to white frost generated by sugar analysis because the fruit and vegetable branches to be detected are positioned at the bottom of the fruit and vegetable particles to be detected in the image. The saidThe larger the gradient line, the longer and straighter the gradient, and the larger the particles of the frosting, the greater the sugar analysis.
Fig. 8 is a schematic diagram of an automatic detection system for freshness of fruits and vegetables according to an eighth embodiment of the present invention, where the method includes:
the obtaining module 801 is configured to obtain a gray level image of a fruit and vegetable to be detected, and obtain an edge cluster of the fruit and vegetable to be detected according to the gray level image.
The plumpness module 802 is configured to calculate a coverage of the fruit and vegetable grain edge cluster to be detected, obtain a grain density of the fruit and vegetable grain edge cluster to be detected according to the coverage, and further obtain the plumpness of the fruit and vegetable to be detected according to the grain density.
The sugar analysis degree module 803 is configured to obtain a to-be-detected fruit and vegetable granule pixel point in the to-be-detected fruit and vegetable granule edge cluster, select and calculate a loss value of the to-be-detected fruit and vegetable granule pixel point, and calculate a sugar analysis degree of the to-be-detected fruit and vegetable according to the to-be-detected fruit and vegetable granule pixel point corresponding to the smallest loss value.
The freshness module 804 is configured to obtain freshness of the fruit and vegetable to be detected according to the fullness and the sugar analysis degree, where the freshness is lower than a preset freshness threshold, and the freshness of the fruit and vegetable to be detected is not qualified.
The technical features and technical effects of the automatic detection system for fruit and vegetable freshness provided by the embodiment of the invention are the same as those of the method provided by the embodiment of the invention, and are not repeated here.
The invention has the following beneficial effects:
Firstly, a gray level image of fruits and vegetables to be detected is obtained, and edge clusters of the fruits and vegetables to be detected are obtained according to the gray level image. Each fruit and vegetable grain edge cluster to be detected represents one fruit and vegetable grain. Secondly, calculating the covering degree of the fruit and vegetable grain edge clusters to be detected, acquiring the grain density of the fruit and vegetable grain edge clusters to be detected according to the covering degree, and further acquiring the plumpness of the fruit and vegetable to be detected according to the grain density. The fullness is a key index for detecting the freshness of fruits and vegetables, and the freshness of fruits and vegetables is higher.
And then, obtaining the pixel points of the fruit and vegetable particles to be detected in the edge cluster of the fruit and vegetable particles to be detected, selecting and calculating the loss value of the pixel points of the fruit and vegetable particles to be detected, and calculating the sugar analysis degree of the fruit and vegetable particles to be detected according to the pixel points of the fruit and vegetable particles to be detected corresponding to the minimum loss value. The sugar analysis degree is also a key index for detecting the freshness of fruits and vegetables, and the more fresh the fruits and vegetables are, the lower the sugar analysis degree is. And finally, acquiring the freshness of the fruits and vegetables to be detected according to the plumpness and the sugar analysis degree, wherein when the freshness is lower than a preset freshness threshold value, the freshness of the fruits and vegetables to be detected is unqualified. By adopting the method, when slight unreliability of fruits and vegetables such as water loss or sugar analysis occurs, the unqualified freshness of the fruits and vegetables can still be detected, so that the freshness detection of the fruits and vegetables is more accurate.
It should be noted that the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (8)
1. The automatic detection method for the freshness of the fruits and vegetables is characterized by comprising the following steps of:
acquiring a gray level image of fruits and vegetables to be detected, and acquiring edge clusters of the fruits and vegetables to be detected according to the gray level image;
Calculating the covering degree of the fruit and vegetable grain edge clusters to be detected, acquiring the grain density of the fruit and vegetable grain edge clusters to be detected according to the covering degree, and further acquiring the plumpness of the fruits and vegetables to be detected according to the grain density;
The method comprises the steps of obtaining a to-be-detected fruit and vegetable particle pixel point in a to-be-detected fruit and vegetable particle edge cluster, selecting and calculating a loss value of the to-be-detected fruit and vegetable particle pixel point, calculating a sugar analysis degree of the to-be-detected fruit and vegetable according to the to-be-detected fruit and vegetable particle pixel point corresponding to the smallest loss value, marking the white point pixel point corresponding to the smallest loss value as a white frost pixel point, obtaining a distance average value between the white frost pixel point and the to-be-detected fruit and vegetable particle edge and a gradient value of the white frost pixel point;
Acquiring the freshness of the fruits and vegetables to be detected according to the fullness and the sugar analysis degree, and if the freshness is lower than a preset freshness threshold value, failing the freshness of the fruits and vegetables to be detected;
the step of obtaining the edge cluster of the fruit and vegetable to be detected according to the gray level graph comprises the following steps:
performing edge detection on the gray level image to obtain an edge image, and obtaining a fruit and vegetable connected domain to be detected according to the edge image;
Carrying out circle fitting on the edges of fruit and vegetable particles to be detected in the fruit and vegetable communication domain to be detected by adopting a least square method to obtain a fitting circle, and obtaining a fitting circle center and a fitting radius of the fitting circle;
searching for the edge of the target fruit and vegetable to be detected, the distance between the edge and the corresponding fitting circle center of which is smaller than a preset first distance value, and screening out the edge of the fruit and vegetable to be detected, corresponding to the fitting radius, of which the absolute value of the difference value between the edge and the target fitting radius is larger than a preset second distance value, so as to obtain a co-circle edge, wherein the fitting radius corresponding to the edge of the target fruit and vegetable to be detected is the target fitting radius;
Calculating a first circle fitting loss value of the edge of the fruit and vegetable particle to be detected and a second circle fitting loss value which is common to the edge of the fruit and vegetable particle to be detected and the corresponding co-circle edge, acquiring an edge coefficient according to the first circle fitting loss value and the second circle fitting loss value, and when the edge coefficient is larger than a preset edge threshold value, the edge of the fruit and vegetable particle to be detected is a fruit and vegetable particle edge cluster.
2. The automated fruit and vegetable freshness detection method according to claim 1, wherein obtaining the edge coefficients from the first circle fit loss value and the second circle fit loss value comprises:
the inverse of the first circle fit loss value multiplied by the second circle fit loss value is used as the input value of the base e exponential function, and the output value of the exponential function is used as the edge coefficient.
3. The automated fruit and vegetable freshness detection method according to claim 1, wherein obtaining the coverage of the edge clusters of the fruit and vegetable to be detected comprises:
Obtaining a fitting radius and an edge length sum of an edge cluster of the fruit and vegetable to be detected;
And obtaining a fitting circle circumference according to the fitting radius, wherein the difference value of the fitting circle circumference minus the edge length sum is used as an input value of a hyperbolic function, and an output value of the hyperbolic function is used as the covering degree of the edge cluster of the fruit and vegetable to be detected.
4. The automated fruit and vegetable freshness detection method according to claim 1, wherein the obtaining of the particle density of the fruit and vegetable edge clusters to be detected according to the degree of coverage comprises:
multiplying the area of a fitting circle corresponding to the fruit and vegetable particles to be detected by the coverage corresponding to the fruit and vegetable particles to be detected to obtain a first product;
And repeating the first product obtaining process to obtain first products corresponding to the fruit and vegetable particles to be detected, sequentially adding the first products, and subtracting the area of the connected areas of the fruit and vegetable particles to be detected to obtain the particle density of the edge clusters of the fruit and vegetable particles to be detected.
5. The automated fruit and vegetable freshness detection method according to claim 1, wherein obtaining the fullness of the fruit and vegetable to be detected according to the particle density comprises:
Obtaining a loss value of a fitting circle of the fruit and vegetable grain edge cluster to be detected and a minimum distance between the fruit and vegetable grain edge cluster to be detected and a surrounding edge cluster, wherein the value obtained by dividing the loss value by the minimum distance is used as a first ratio;
Repeating the first ratio obtaining process to obtain a first ratio corresponding to each fruit and vegetable grain edge cluster to be detected, and obtaining an average value of the first ratios;
And taking the average value as an input value of a hyperbolic function, and multiplying the output value of the hyperbolic function by the particle density to obtain the plumpness of the fruits and vegetables to be detected.
6. The automated fruit and vegetable freshness detection method according to claim 1, wherein selecting and calculating a loss value of pixels of fruit and vegetable particles to be detected comprises:
acquiring the gray average value of the pixels of the fruit and vegetable particles to be detected in the edge cluster of the fruit and vegetable particles to be detected, acquiring the difference value between the gray value and the gray average value of each pixel of the fruit and vegetable particles to be detected, and sequencing the pixels of the fruit and vegetable particles to be detected according to the difference value from large to small;
Selecting any number of fruit and vegetable particle pixels to be detected which are ranked forward as white point pixels, and selecting any number of fruit and vegetable particle pixels to be detected which are ranked forward as light spot pixels;
the vector between the spot pixel point and the circle center of the corresponding fitting circle is recorded as a spot vector;
Dividing the variance of the gray value of the white point pixel point by the information entropy of the light spot vector to obtain a second ratio, and multiplying the opposite number of the second ratio by the gray average value of the white point pixel point to obtain an input value of an exponential function based on e, wherein the output value of the exponential function is used as the loss value of the pixel point of the fruit and vegetable particles to be detected.
7. The automated fruit and vegetable freshness detection method according to claim 1, wherein obtaining the freshness of the fruit and vegetable to be detected according to the fullness and sugar analysis degree comprises:
and dividing the fullness by the sugar analysis degree to obtain the freshness of the fruits and vegetables to be detected.
8. Be applied to automatic detecting system of fruit vegetables freshness, its characterized in that includes:
The acquisition module is used for acquiring a gray level image of fruits and vegetables to be detected, and acquiring edge clusters of the fruits and vegetables to be detected according to the gray level image;
The plumpness module is used for calculating the covering degree of the edge cluster of the fruit and vegetable grains to be detected, acquiring the grain density of the edge cluster of the fruit and vegetable grains to be detected according to the covering degree, and further acquiring the plumpness of the fruit and vegetable grains to be detected according to the grain density;
the sugar analysis degree module is used for acquiring to-be-detected fruit and vegetable particle pixel points in the to-be-detected fruit and vegetable particle edge cluster, selecting and calculating a loss value of the to-be-detected fruit and vegetable particle pixel points, and calculating the sugar analysis degree of the to-be-detected fruit and vegetable according to the to-be-detected fruit and vegetable particle pixel points corresponding to the smallest loss value, wherein the sugar analysis degree module comprises the steps of recording white point pixel points corresponding to the smallest loss value as white frost pixel points, acquiring a distance average value between the white frost pixel points and the to-be-detected fruit and vegetable particle edge and a gradient value of the white frost pixel points;
The freshness module is used for acquiring freshness of the fruits and vegetables to be detected according to the fullness and the sugar analysis degree, and when the freshness is lower than a preset freshness threshold value, the freshness of the fruits and vegetables to be detected is unqualified;
the step of obtaining the edge cluster of the fruit and vegetable to be detected according to the gray level graph comprises the following steps:
performing edge detection on the gray level image to obtain an edge image, and obtaining a fruit and vegetable connected domain to be detected according to the edge image;
Carrying out circle fitting on the edges of fruit and vegetable particles to be detected in the fruit and vegetable communication domain to be detected by adopting a least square method to obtain a fitting circle, and obtaining a fitting circle center and a fitting radius of the fitting circle;
searching for the edge of the target fruit and vegetable to be detected, the distance between the edge and the corresponding fitting circle center of which is smaller than a preset first distance value, and screening out the edge of the fruit and vegetable to be detected, corresponding to the fitting radius, of which the absolute value of the difference value between the edge and the target fitting radius is larger than a preset second distance value, so as to obtain a co-circle edge, wherein the fitting radius corresponding to the edge of the target fruit and vegetable to be detected is the target fitting radius;
Calculating a first circle fitting loss value of the edge of the fruit and vegetable particle to be detected and a second circle fitting loss value which is common to the edge of the fruit and vegetable particle to be detected and the corresponding co-circle edge, acquiring an edge coefficient according to the first circle fitting loss value and the second circle fitting loss value, and when the edge coefficient is larger than a preset edge threshold value, the edge of the fruit and vegetable particle to be detected is a fruit and vegetable particle edge cluster.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510152196.4A CN119666666B (en) | 2025-02-12 | 2025-02-12 | Automatic detection method and system applied to freshness of fruits and vegetables |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510152196.4A CN119666666B (en) | 2025-02-12 | 2025-02-12 | Automatic detection method and system applied to freshness of fruits and vegetables |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119666666A CN119666666A (en) | 2025-03-21 |
| CN119666666B true CN119666666B (en) | 2025-09-23 |
Family
ID=95002334
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510152196.4A Active CN119666666B (en) | 2025-02-12 | 2025-02-12 | Automatic detection method and system applied to freshness of fruits and vegetables |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119666666B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118130743A (en) * | 2024-05-08 | 2024-06-04 | 冠珹餐饮管理(沈阳)有限公司 | Water quality monitoring data processing method and system based on cloud computing |
| CN118392866A (en) * | 2024-03-26 | 2024-07-26 | 深圳个元科技有限公司 | A method and system for automatic visual inspection of commutator appearance defects |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190043193A1 (en) * | 2017-08-01 | 2019-02-07 | Retina-Ai Llc | Systems and Methods Using Weighted-Ensemble Supervised-Learning for Automatic Detection of Retinal Disease from Tomograms |
| CN118485655A (en) * | 2024-05-29 | 2024-08-13 | 依未科技(北京)有限公司 | Image processing method and device, readable storage medium and electronic equipment |
-
2025
- 2025-02-12 CN CN202510152196.4A patent/CN119666666B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118392866A (en) * | 2024-03-26 | 2024-07-26 | 深圳个元科技有限公司 | A method and system for automatic visual inspection of commutator appearance defects |
| CN118130743A (en) * | 2024-05-08 | 2024-06-04 | 冠珹餐饮管理(沈阳)有限公司 | Water quality monitoring data processing method and system based on cloud computing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119666666A (en) | 2025-03-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Blasco et al. | Computer vision detection of peel defects in citrus by means of a region oriented segmentation algorithm | |
| Diaz et al. | Comparison of three algorithms in the classification of table olives by means of computer vision | |
| Brosnan et al. | Inspection and grading of agricultural and food products by computer vision systems—a review | |
| Xu et al. | Identification of freezer burn on frozen salmon surface using hyperspectral imaging and computer vision combined with machine learning algorithm | |
| Luzuriaga et al. | Analysis of visual quality attributes of white shrimp by machine vision | |
| Markovic et al. | Color measurement of food products using CIE L* a* b* and RGB color space. | |
| Jhawar | Orange sorting by applying pattern recognition on colour image | |
| Quevedo et al. | Determination of senescent spotting in banana (Musa cavendish) using fractal texture Fourier image | |
| Thinh et al. | Mango classification system based on machine vision and artificial intelligence | |
| Koszela et al. | Computer image analysis in the quality in procedure for selected carrot varieties | |
| Chopde et al. | Developments in computer vision system, focusing on its applications in quality inspection of fruits and vegetables-A review | |
| Hongwiangjan et al. | Evaluation of pomelo maturity based on acoustic response and peel properties | |
| Raji et al. | Prospects of Cumputer Vision Automated Sorting Systems in Agricultural Process Operations in Nigeria | |
| Ahmad et al. | Color distribution analysis for ripeness prediction of golden apollo melon | |
| Donis-González et al. | Color vision system to assess English walnut (Juglans Regia) kernel pellicle color | |
| Bakar et al. | Development of ripeness indicator for quality assessment of harumanis mango by using image processing technique | |
| CN119666666B (en) | Automatic detection method and system applied to freshness of fruits and vegetables | |
| Novas et al. | Development of a smartphone application for assessment of chilling injuries in zucchini | |
| KR102596206B1 (en) | Classification system of dried sea cucumber using image analysis | |
| JP2010227892A (en) | Foreign matter sorting method and foreign matter sorting equipment | |
| Hafezi et al. | The effect of drying kinetic on shrinkage and colour of potato slices in the vacuum-infrared drying method | |
| Díaz | Classification and quality evaluation of table olives | |
| Matiacevich et al. | Quality assessment of blueberries by computer vision | |
| Lee et al. | Color quantization and image analysis for automated fruit quality evaluation | |
| Riyadi et al. | Optimized estimation of mangosteen maturity stage using svm and color features combination approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |