CN112507967A - Image processing system based on artificial intelligence recognition - Google Patents
Image processing system based on artificial intelligence recognition Download PDFInfo
- Publication number
- CN112507967A CN112507967A CN202011543214.5A CN202011543214A CN112507967A CN 112507967 A CN112507967 A CN 112507967A CN 202011543214 A CN202011543214 A CN 202011543214A CN 112507967 A CN112507967 A CN 112507967A
- Authority
- CN
- China
- Prior art keywords
- face
- image
- recognition
- information
- human body
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Molecular Biology (AREA)
- Evolutionary Biology (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a face recognition system, in particular to an image processing system based on artificial intelligence recognition, which comprises: the human body recognition module is used for realizing the acquisition of image depth information and realizing the recognition of human body information in the image data according to the image depth information based on the fuzzy depth neural network model; the dangerous behavior recognition module is started when a human body is recognized, is used for acquiring human body depth information and skeleton information and recognizing dangerous behaviors according to all the skeleton information based on an infinite depth neural network; the target image mining module is started when dangerous behaviors are found, and is used for mining a target face image set based on a Dssd-inclusion-V3 coco model; and the face recognition module is used for extracting the LBPH, SIFT and face skeleton characteristics of the face image set and realizing the face recognition based on a decision tree. The invention can realize the timely capture and recognition of the face image.
Description
Technical Field
The invention relates to a face recognition system, in particular to an image processing system based on artificial intelligence recognition.
Background
At present, video image monitoring in a monitoring area is an important technical field which is continuously developed and researched by a plurality of technicians related to video monitoring. Because if the method completely depends on manual monitoring, the method has the practical defects of low efficiency, increased cost and high monitoring error rate or leakage rate. To solve these drawbacks, many persons skilled in the relevant art have conducted intensive research and development.
At present, human face recognition is added in video monitoring for real-time monitoring, which is a widely used technical means, but capturing human faces at a certain distance is an important technical problem. Because the human face can be presented and captured only under a certain close distance, the possible target object is very close to the protected object to be protected, such as cultural relic protection, exhibition protection scenes and the like. In these scenes, due to factors such as lighting, image monitoring is not necessarily timely for capturing human body, which results in a problem of delay in capturing subsequent human face, and this may cause a great risk and a negative effect on the actual protection effect of the protected object in these scenes.
Disclosure of Invention
In order to solve the technical problems, the invention provides an image processing system based on artificial intelligence recognition, which can realize the timely capture and recognition of face images.
In order to achieve the purpose, the invention adopts the technical scheme that:
an image processing system based on artificial intelligence recognition, comprising:
the human body recognition module is used for realizing the acquisition of image depth information and realizing the recognition of human body information in the image data according to the image depth information based on the fuzzy depth neural network model;
the dangerous behavior recognition module is started when a human body is recognized, is used for acquiring human body depth information and skeleton information and recognizing dangerous behaviors according to all the skeleton information based on an infinite depth neural network;
the target image mining module is started when dangerous behaviors are found, and is used for mining a target face image set based on a Dssd-inclusion-V3 coco model;
and the face recognition module is used for extracting the LBPH, SIFT and face skeleton characteristics of the face image set and realizing the face recognition based on a decision tree.
Further, still include:
and the image preprocessing module is used for determining the deflection angle of each video frame according to the coordinate information of each video frame and reconstructing other video frames according to the deflection angle of one video frame.
Furthermore, the dangerous behavior identification module realizes acquisition of human body depth information and skeleton information based on a kinect depth sensor, and eliminates jitter and noise interference of lock acquisition skeleton information based on a skeleton information filtering module.
Further, the Dssd-inclusion-V3 coco model adopts a Dssd target detection algorithm, the inclusion-V3 deep neural network is pre-trained by a coco data set, then the model is trained by a previously prepared data set, various parameters in the deep neural network are finely adjusted, and finally a suitable target detection model for the face image is obtained.
Furthermore, the human body recognition module firstly calls a video frame taking script, obtains an image at a certain frame number, then realizes the acquisition of image depth information based on a kinect depth sensor, and realizes the recognition of human body information in image data according to the image depth information based on a fuzzy depth neural network model.
Further, the face recognition module realizes face recognition according to LBPH, SIFT and face bone features of the face image set based on a CART algorithm, wherein the CART algorithm is obtained based on training of the LBPH, SIFT and face bone features of the safe face image set.
Further, the face recognition module firstly realizes the recognition of the face image set on the basis of a Dssd-inclusion-V3 coco model, directly warns if a barrier is found, extracts the LBPH, SIFT and face skeleton characteristics of the face image set if no barrier is found, and then realizes the recognition of the face on the basis of a decision tree.
Further, the reconstruction of other video frames is completed through the following steps;
calculating a supplementary deflection angle of each video frame according to the deflection angle of each video frame;
redrawing each video frame according to its supplemental deflection angle.
The invention has the following beneficial effects:
1) the human face image set can be captured in time based on the design of the human body recognition module, the dangerous behavior recognition module and the target image mining module;
2) based on the adjustment of the deflection angle of the video frame, the behavior and the face recognition error caused by the difference of the video frame angles can be well avoided;
3) the LBPH, SIFT and human face skeleton characteristics are adopted for face recognition, the influence of single characteristics on recognition due to illumination, angle and scale change can be effectively improved, and the recognition rate is improved.
4) The infinite deep neural network is adopted to realize the identification of dangerous behaviors according to all skeleton information, so that the rapid identification of dynamic dangerous behaviors can be realized, and the real-time performance is good.
Drawings
Fig. 1 is a system block diagram of an image processing system based on artificial intelligence recognition according to an embodiment of the present invention.
Fig. 2 is a flowchart illustrating an image processing system based on artificial intelligence recognition according to an embodiment of the present invention.
Detailed Description
In order that the objects and advantages of the invention will be more clearly understood, the invention is further described in detail below with reference to examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Example 1
As shown in fig. 1, an embodiment of the present invention provides an image processing system based on artificial intelligence recognition, including:
the image preprocessing module is used for determining the deflection angle of each video frame according to the coordinate information of each video frame and reconstructing other video frames according to the deflection angle of one video frame;
the human body recognition module is used for realizing the acquisition of image depth information and realizing the recognition of human body information in the image data according to the image depth information based on the fuzzy depth neural network model;
the dangerous behavior recognition module is started when a human body is recognized, is used for acquiring human body depth information and skeleton information and recognizing dangerous behaviors according to all the skeleton information based on an infinite depth neural network;
the target image mining module is started when dangerous behaviors are found, and is used for mining a target face image set based on a Dssd-inclusion-V3 coco model;
and the face recognition module is used for extracting the LBPH, SIFT and face skeleton characteristics of the face image set and realizing the face recognition based on a decision tree.
In this embodiment, the dangerous behavior recognition module realizes human depth information and bone information acquisition based on a kinect depth sensor, and eliminates jitter and noise interference of the lock acquiring bone information based on a bone information filtering module.
In this embodiment, the Dssd _ inclusion _ V3 coco model adopts a Dssd target detection algorithm, a coco data set is used to pre-train the inclusion _ V3 deep neural network, then the model is trained by using a previously prepared data set, each parameter in the deep neural network is finely tuned, and finally a suitable target detection model for a face image is obtained.
In this embodiment, the human body identification module first calls a video frame taking script, obtains an image at intervals of a certain number of frames, then obtains image depth information based on a kinect depth sensor, and identifies human body information in image data according to the image depth information based on a fuzzy depth neural network model.
In this embodiment, the face recognition module recognizes a face according to the LBPH, SIFT, and facial bone features of a face image set based on a CART algorithm, where the CART algorithm is obtained by training based on the LBPH, SIFT, and facial bone features of a safe face image set.
In this embodiment, the face recognition module first identifies an obstruction in a face image set based on a Dssd _ inclusion _ V3 coco model, directly warns if the obstruction is found, extracts LBPH, SIFT and face skeleton features of the face image set if the obstruction is not found, and then identifies a face based on a decision tree.
In this embodiment, the reconstruction of other video frames is completed through the following steps;
calculating a supplementary deflection angle of each video frame according to the deflection angle of each video frame;
redrawing each video frame according to its supplemental deflection angle.
As shown in fig. 2, the present embodiment includes the following steps:
s1, determining the deflection angle of each video frame according to the coordinate information of each video frame, and reconstructing other video frames according to the deflection angle of one video frame;
s2, obtaining image depth information, and recognizing human body information in image data according to the image depth information based on the fuzzy depth neural network model;
s3, starting when a human body is identified, acquiring human body depth information and skeleton information, and identifying dangerous behaviors according to all skeleton information based on an infinite depth neural network;
s4, starting when dangerous behaviors are found, and mining a target face image set based on a Dssd-inclusion-V3 coco model;
s5, extracting LBPH, SIFT and human face skeleton characteristics of the human face image set, and realizing human face recognition based on a decision tree. Specifically, the method comprises the following steps: the face recognition module realizes face recognition according to LBPH, SIFT and face skeleton characteristics of a face image set based on a CART algorithm, wherein the CART algorithm is obtained by training based on LBPH, SIFT and face skeleton characteristics of a safe face image set.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be construed as the protection scope of the present invention.
Claims (8)
1. An image processing system based on artificial intelligence recognition, comprising:
the human body recognition module is used for realizing the acquisition of image depth information and realizing the recognition of human body information in the image data according to the image depth information based on the fuzzy depth neural network model;
the dangerous behavior recognition module is started when a human body is recognized, is used for acquiring human body depth information and skeleton information and recognizing dangerous behaviors according to all the skeleton information based on an infinite depth neural network;
the target image mining module is started when dangerous behaviors are found, and is used for mining a target face image set based on a Dssd-inclusion-V3 coco model;
and the face recognition module is used for extracting the LBPH, SIFT and face skeleton characteristics of the face image set and realizing the face recognition based on a decision tree.
2. The image processing system based on artificial intelligence recognition of claim 1, further comprising:
and the image preprocessing module is used for determining the deflection angle of each video frame according to the coordinate information of each video frame and reconstructing other video frames according to the deflection angle of one video frame.
3. The image processing system based on artificial intelligence recognition of claim 1, wherein the dangerous behavior recognition module realizes human depth information and bone information acquisition based on a kinect depth sensor, and eliminates jitter and noise interference of bone information acquisition based on a bone information filtering module.
4. The artificial intelligence recognition-based image processing system as claimed in claim 1, wherein the Dssd _ inclusion _ V3 coco model employs a Dssd target detection algorithm, the inclusion _ V3 deep neural network is pre-trained with a coco data set, then the model is trained with a previously prepared data set, parameters in the deep neural network are fine-tuned, and finally a suitable target detection model for the face image is obtained.
5. The image processing system as claimed in claim 1, wherein the human body recognition module first calls a video framing script to acquire an image every certain number of frames, then realizes acquisition of image depth information based on a kinect depth sensor, and realizes recognition of human body information in the image data according to the image depth information based on the fuzzy depth neural network model.
6. The image processing system based on artificial intelligence recognition as claimed in claim 1, wherein the face recognition module realizes the recognition of the face according to LBPH, SIFT and facial bone features of the face image set based on CART algorithm, wherein the CART algorithm is trained based on LBPH, SIFT and facial bone features of the safe face image set.
7. The image processing system based on artificial intelligence recognition as claimed in claim 1, wherein the face recognition module firstly recognizes an occlusion in the face image set based on a Dssd _ inclusion _ V3 coco model, directly warns if the occlusion is found, extracts LBPH, SIFT and facial bone features of the face image set if the occlusion is not found, and then recognizes a face based on a decision tree.
8. The image processing system based on artificial intelligence recognition of claim 1, wherein the reconstruction of other video frames is accomplished by the following steps;
calculating a supplementary deflection angle of each video frame according to the deflection angle of each video frame;
redrawing each video frame according to its supplemental deflection angle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011543214.5A CN112507967A (en) | 2020-12-23 | 2020-12-23 | Image processing system based on artificial intelligence recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011543214.5A CN112507967A (en) | 2020-12-23 | 2020-12-23 | Image processing system based on artificial intelligence recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112507967A true CN112507967A (en) | 2021-03-16 |
Family
ID=74923293
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011543214.5A Pending CN112507967A (en) | 2020-12-23 | 2020-12-23 | Image processing system based on artificial intelligence recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112507967A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114266998A (en) * | 2021-12-24 | 2022-04-01 | 南京米保网络科技有限公司 | A method using sensors to reduce risk in woodworking machinery |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103136899A (en) * | 2013-01-23 | 2013-06-05 | 宁凯 | Intelligent alarming monitoring method based on Kinect somatosensory equipment |
| CN111325133A (en) * | 2020-02-17 | 2020-06-23 | 深圳龙安电力科技有限公司 | Image processing system based on artificial intelligence recognition |
| CN111368803A (en) * | 2020-03-28 | 2020-07-03 | 河南工业职业技术学院 | A face recognition method and system |
| CN111414886A (en) * | 2020-03-28 | 2020-07-14 | 福建工程学院 | Intelligent recognition system for human body dynamic characteristics |
| CN112085003A (en) * | 2020-09-24 | 2020-12-15 | 湖北科技学院 | Method and device for automatic identification of abnormal behavior in public places, and camera equipment |
-
2020
- 2020-12-23 CN CN202011543214.5A patent/CN112507967A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103136899A (en) * | 2013-01-23 | 2013-06-05 | 宁凯 | Intelligent alarming monitoring method based on Kinect somatosensory equipment |
| CN111325133A (en) * | 2020-02-17 | 2020-06-23 | 深圳龙安电力科技有限公司 | Image processing system based on artificial intelligence recognition |
| CN111368803A (en) * | 2020-03-28 | 2020-07-03 | 河南工业职业技术学院 | A face recognition method and system |
| CN111414886A (en) * | 2020-03-28 | 2020-07-14 | 福建工程学院 | Intelligent recognition system for human body dynamic characteristics |
| CN112085003A (en) * | 2020-09-24 | 2020-12-15 | 湖北科技学院 | Method and device for automatic identification of abnormal behavior in public places, and camera equipment |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114266998A (en) * | 2021-12-24 | 2022-04-01 | 南京米保网络科技有限公司 | A method using sensors to reduce risk in woodworking machinery |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112487921B (en) | Face image preprocessing method and system for living body detection | |
| Singh et al. | Muhavi: A multicamera human action video dataset for the evaluation of action recognition methods | |
| CN109145803B (en) | Gesture recognition method and device, electronic equipment and computer readable storage medium | |
| CN108446617A (en) | The human face quick detection method of anti-side face interference | |
| CN110853295A (en) | High-altitude parabolic early warning method and device | |
| CN108446690B (en) | Human face in-vivo detection method based on multi-view dynamic features | |
| CN110084130B (en) | Face screening method, device, equipment and storage medium based on multi-target tracking | |
| WO2014092552A2 (en) | Method for non-static foreground feature extraction and classification | |
| CN112613430B (en) | Gait recognition method based on deep migration learning | |
| CN115240280B (en) | Method for constructing human face living body detection classification model, detection classification method and device | |
| CN111767788A (en) | Non-interactive monocular in vivo detection method | |
| CN115620405A (en) | Face fake video detection method based on rPPG heart rate characteristics | |
| CN115937237A (en) | Local feature extraction method based on edge transform domain | |
| CN113420582A (en) | Anti-counterfeiting detection method and system for palm vein recognition | |
| CN112507967A (en) | Image processing system based on artificial intelligence recognition | |
| CN110378935A (en) | Parabolic recognition methods based on image, semantic information | |
| CN103426005B (en) | Automatic database creating video sectioning method for automatic recognition of micro-expressions | |
| CN111881818B (en) | Medical action fine-grained recognition device and computer-readable storage medium | |
| CN113255549A (en) | Intelligent recognition method and system for pennisseum hunting behavior state | |
| CN107944424A (en) | Front end human image collecting and Multi-angle human are distributed as comparison method | |
| CN117671584A (en) | Method, device, equipment, medium and product for detecting personnel gathering area | |
| CN117456586A (en) | Micro expression recognition method, system, equipment and medium | |
| CN111414886A (en) | Intelligent recognition system for human body dynamic characteristics | |
| CN107403192B (en) | Multi-classifier-based rapid target detection method and system | |
| Bal et al. | Plate Number Recognition Using Segmented Method With Artificial Neural Network |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210316 |
|
| RJ01 | Rejection of invention patent application after publication |