TWI836968B - Method for making customized 3d image and system thereof - Google Patents

Abstract

The present invention discloses a method for making customized 3D image and a system thereof. The method comprises capturing a first raw image information and a second raw image information by using a capturing unit; identifying the first raw image information and generating a plurality of first partial images according thereto, and identifying the second raw image information and generating a plurality of second partial images according thereto by using an identifying unit; computing a default pose 3D information and generating a pose angle information and a pose part information by using a computing processing unit; and recombining the first partial images and the second partial images referring to the pose angle information and the pose part information so as to generate a 3D image model corresponding thereto. The 3D image model is then provided for 3D object printing so that the purpose of customization and complete preservation is realized.

Description

Customized stereoscopic image production method and system

The present invention relates to an image production technology, and in particular, to a customized three-dimensional image production method and system.

When most pregnant women are pregnant, in order to monitor the growth of the fetus, they often use ultrasound to understand the actual growth. Ultrasound is an image created based on the principle of sound wave reflection. When the emitted ultrasound signal hits an object, it will produce reflection and refraction. , since the reflectivity and refractive index of human tissues are different, the ultrasonic scanning instrument will construct the position, size, shape, structure, etc. of the mother's uterus and fetus based on the received ultrasonic signal reflection.

The most common ultrasound scanners are divided into 2D, 3D and 4D. 2D ultrasound is a black and white flat image, while 3D stereo ultrasound uses a computer image processing system to reassemble multiple 2D pictures into a three-dimensional image. 4D dynamic stereo ultrasound is a dynamic three-dimensional image, which uses a large number of 3D images to form a continuous dynamic video, which can clearly show the movement of the fetus. Generally, ultrasound examinations are performed at about 7-8 weeks, 12 weeks, and 20-24 weeks. Such precious images can mostly only be preserved in the form of photos, which are stored in a single way. In addition, the photos printed out by general ultrasound examinations are printed on thermal paper. If heated or left for too long, the clarity will be affected and become increasingly blurred.

Furthermore, no matter what kind of ultrasound image it is, it may be because the fetus stretches or moves in the womb, causing the limbs to cover the face, or the placenta or other intrauterine substances may obscure the limbs, resulting in the inability to capture a complete ultrasound of the fetus. Sonic images, even dynamic three-dimensional images, may encounter this problem. As a result, even if the images are obtained, the required commemorative items cannot be made. Therefore, how to balance image integrity and customization is an urgent problem to be solved. problem.

The main purpose of the present invention is to provide a customized stereoscopic image production method, which recognizes the original image information to generate multiple partial images, and then reassembles the partial images with the preset posture stereoscopic image information to obtain a complete stereoscopic image model.

Another object of the present invention is to provide a customized stereoscopic image production system, in which the capture unit obtains the original image information, and after the recognition unit recognizes it, the calculation processing unit reorganizes the recognized part of the image according to the needs to achieve the purpose of customization.

In order to achieve the above-mentioned purpose, an embodiment of the present invention discloses a customized stereoscopic image production method, the steps of which include: capturing a first original image information and a second original image information; identifying the first original image information to generate a corresponding plurality of first partial images, and identifying the second original image information to generate a corresponding plurality of second partial images; calculating a preset posture stereoscopic image information to generate a corresponding posture angle information and a posture part information; and recombining the first partial images and the second partial images with the posture angle information and the posture part information to generate a corresponding stereoscopic image model.

Preferably, in the steps of identifying the first original image information and generating a plurality of corresponding first partial images, and identifying the second original image information and generating a plurality of corresponding second partial images, the third An original image information, generate a corresponding first image pixel value matrix, and convert the second original image information, generate a corresponding second image pixel value matrix, and filter the first image with a preset image pixel value matrix The noise of the pixel value matrix and the second image pixel value matrix generates a first processed image information and a second processed image information, identifies the first processed image information, and generates the corresponding first partial images, and identifying the second processed image information to generate the corresponding second partial images.

Preferably, in the step of recombining the first partial images and the second partial images with the posture angle information and the posture part information to generate a corresponding stereoscopic image model, the posture angle information and the posture part information are used to correspond to the first partial images and the second partial images, and the corresponding plurality of image components are selected, and the image components are recombined to generate the corresponding stereoscopic image model.

Preferably, after the step of recombining the first partial images and the second partial images using the posture angle information and the posture part information to generate a corresponding three-dimensional image model, the steps include: Print the three-dimensional image model to generate a corresponding three-dimensional object.

Preferably, the first original image information and the second original image information are ultrasound images.

In order to achieve another of the above objectives, one embodiment of the present invention discloses a customized three-dimensional image production system, including: a capture unit that captures first original image information and a second original image of a human body Information; a recognition unit, signal-connected to the capture unit, receives and recognizes the first original image information, generates a plurality of corresponding first partial images, and receives and recognizes the second original image information, generates a corresponding plurality of a second partial image; and an arithmetic processing unit, signal-connected to the recognition unit, to perform calculations with a preset attitude stereoscopic image information, to generate and combine the first parts with an attitude angle information and an attitude part information. The partial image is recombined with the second partial images to generate a corresponding three-dimensional image model.

Preferably, the operation processing unit converts the first original image information to generate a corresponding first image pixel value matrix, and converts the second original image information to generate a corresponding second image pixel value matrix, and filters the noise of the first image pixel value matrix and the second image pixel value matrix with a preset image pixel value matrix, generates and transmits a first processed image information and a second processed image information to the recognition unit to recognize the first processed image information to generate the corresponding first partial images, and recognize the second processed image information to generate the corresponding second partial images.

Preferably, the computing unit uses the posture angle information and the posture part information to select corresponding plurality of image components among the first partial images and the second partial images, and use The image components are reorganized to generate the corresponding three-dimensional image model.

Preferably, a three-dimensional printing device is included, which is connected to the signal of the operation processing unit, receives and generates a corresponding three-dimensional object according to the three-dimensional image model.

Preferably, the capturing unit is an ultrasonic scanner.

The beneficial effect of the present invention is that it takes into account the integrity of the image according to different needs, and produces commemorative or practical cultural and creative products with customized three-dimensional images.

The relevant patented features and technical content of the present invention will be clearly presented in the following detailed description of the preferred embodiments with reference to the drawings.

Please refer to the first figure, which is a method flow chart according to an embodiment of the present invention. As shown in the figure, this method consists of the following steps.

Step S1: Capture a first original image information and a second original image information;

Step S2: identifying the first original image information to generate a plurality of corresponding first partial images, and identifying the second original image information to generate a plurality of corresponding second partial images;

Step S3: Calculate a preset posture stereoscopic image information to generate corresponding posture angle information and posture part information; and

Step S4: Recombining the first partial images and the second partial images with the posture angle information and the posture part information to generate a corresponding three-dimensional image model.

Please also refer to the second figure, which is a schematic diagram of a system of an embodiment of the present invention. As shown in the figure, the customized stereoscopic image production system includes: a capture unit 1, a recognition unit 2, an operation processing unit 3 and a stereoscopic printing device 4, wherein the capture unit 1 is connected to the recognition unit 2 by signal, and the operation processing unit 3 is connected to the recognition unit 2 and the stereoscopic printing device 4 by signal, respectively.

The capture unit 1 is used to capture the first original image information and the second original image information of the human body. In one embodiment, the capture unit 1 is an ultrasonic scanner, and the first original image information and the second original image information are The image information is an ultrasound image. The first original image information and the second original image information take fetal images as an example. The ultrasound image can be a two-dimensional image or a three-dimensional image, but is not limited to this.

The recognition unit 2 recognizes the first original image information and the second original image information to respectively generate a plurality of first partial images and a plurality of second partial images. The partial images are various limb parts of the fetus image, but are not limited thereto.

The calculation processing unit 3 calculates the preset posture stereoscopic image information to generate corresponding posture angle information and posture part information, and selects corresponding multiple image components from the first part images and the second part images based on the posture angle information and the posture part information, and reorganizes the image components to generate corresponding stereoscopic image models, for example: reorganizing the first part images and the second part images into a stereoscopic image model of the fetus's upper limbs in an embracing posture, but not limited to this.

In one embodiment, when the acquired original image information contains occluded parts or noise, the arithmetic processing unit 3 is used to convert the first original image information into the corresponding first image pixel value matrix, and convert the second original image information. is the corresponding second image pixel value matrix, and uses the preset image pixel value matrix to filter out noise in the first image pixel value matrix and the second image pixel value matrix, and generates and transmits the first processed image information and the second processed image information. The image information is then transmitted to the recognition unit 2, and the first processed image information and the second processed image information are further used for recognition, but this is not limited thereto.

The three-dimensional printing device 4 receives and prints the three-dimensional image model to obtain the corresponding three-dimensional object. In one embodiment, the printing method of the three-dimensional printing device 4 can be extrusion type, metal line type, particle type, or powder spray type. Ink needle type, lamination type or photopolymerization type, but not limited to this.

In one embodiment, it may further include a display interface (not shown), which is connected to the computing processing unit 3 through signals. The user can select the required preset attitude stereoscopic image and the type of three-dimensional object to be made through the display interface. Preferably, multiple images from different periods can be selected through the display interface and combined into a three-dimensional image model, but this is not a limitation.

As described in step S1, the capturing unit 1 captures the first original image information and the second original image information. The amount of the original image information is not limited. Meanwhile, the original image information is not limited to static images but can also be dynamic images, but is not limited thereto.

As described in step S2, the first original image information and the second original image information are identified by the identification unit 2, and a plurality of first partial images and a plurality of second partial images are respectively generated. In one embodiment, when the When the first original image information and the second original image information have noise, first, the operation processing unit 3 will respectively convert the first original image information and the second original image information into the corresponding first image pixel value matrix and the second image pixel value. matrix, and then use the preset image pixel value matrix to filter out the noise in the first image pixel value matrix and the second image pixel value matrix, and generate the first processed image information and the second processed image information, and then use the first processed image information and the second processed image information for identification, but this is not limited to this.

As described in step S3, the computing processing unit 3 is used to calculate the posture stereoscopic image information to obtain the corresponding posture angle information and posture part information. The default posture stereoscopic image information can be set according to the user's needs, such as: standing posture, sitting posture, and lying posture. , arms around gestures, etc.

As described in step S4, the posture angle information and posture part information obtained in the previous step correspond to the first partial images and the second partial images, and a plurality of corresponding image components are selected and reassembled with the image components to generate a corresponding three-dimensional image model, thereby realizing customized three-dimensional image production, and the three-dimensional object can be further printed out with the three-dimensional printing device 4.

To sum up, the customized three-dimensional image production method and system according to one embodiment of the present invention recognizes the original image information to generate multiple partial images, and then converts the partial images using the preset posture three-dimensional image information. Part of the image is reorganized to obtain a three-dimensional image model to achieve customization and improve the shortcomings of conventional methods that cannot be preserved for a long time. Therefore, the purpose of the present invention can indeed be achieved.

However, what is described above is only a preferred embodiment of the present invention, and should not be used to limit the scope of implementation of the present invention. That is, all simple equivalent changes and modifications made according to the scope of the patent application of the present invention and the content of the invention description are still within the scope of the present patent.

1: Capture unit 2:Identification unit 3:Arithmetic processing unit 4: Three-dimensional printing device S1: Steps S2: Step S3: Steps S4: Steps

Figure 1: It is a method flow chart of an embodiment of the present invention; and The second figure is a system schematic diagram of an embodiment of the present invention.

S1-S4: Steps

Claims (10)

A customized three-dimensional image production method, the steps include: a capture unit captures a first original image information and a second original image information; a recognition unit recognizes the first original image information and generates a corresponding plurality of third a part of the image, and recognize the second original image information to generate a plurality of corresponding second partial images; a computing unit operates a preset posture stereoscopic image information to generate corresponding posture angle information and posture part information. ; and the processing unit uses the posture angle information and the posture part information to recombine the first partial images and the second partial images to generate a corresponding three-dimensional image model. According to the customized three-dimensional image production method described in claim 1, after identifying the first original image information, generating a plurality of corresponding first partial images, and identifying the second original image information, generating a corresponding plurality of In the step of the second partial image, the first original image information is converted to generate a corresponding first image pixel value matrix, and the second original image information is converted to generate a corresponding second image pixel value matrix, with a The default image pixel value matrix filters out noise from the first image pixel value matrix and the second image pixel value matrix, generates a first processed image information and a second processed image information, and identifies the first processed image information, Generate the corresponding first partial images, identify the second processed image information, and generate the corresponding second partial images. According to the customized stereoscopic image production method described in claim 1, in the step of recombining the first partial images and the second partial images with the posture angle information and the posture part information to generate a corresponding stereoscopic image model, the posture angle information and the posture part information are used to correspond to the first partial images and the second partial images, and a plurality of corresponding image components are selected, and the image components are recombined to generate the corresponding stereoscopic image model. According to the customized stereoscopic image production method described in claim 1, after the step of recombining the first part images and the second part images with the posture angle information and the posture part information to generate a corresponding stereoscopic image model, the step includes: printing the stereoscopic image model to generate a corresponding stereoscopic object. According to the customized stereoscopic image production method described in claim 1, the first original image information and the second original image information are ultrasound images. A customized stereoscopic image production system includes: a capture unit, which captures a first original image information and a second original image information of a human body; a recognition unit, which is connected to the capture unit signal, receives and recognizes the first original image information, generates a plurality of corresponding first partial images, and receives and recognizes the second original image information, generates a plurality of corresponding second partial images; and a calculation processing unit, which is connected to the recognition unit signal, performs calculations with a preset posture stereoscopic image information, generates and recombines the first partial images and the second partial images with a posture angle information and a posture part information, and generates a corresponding stereoscopic image model. According to the customized stereoscopic image production system described in claim 6, the operation processing unit converts the first original image information to generate a corresponding first image pixel value matrix, and converts the second original image information to generate a corresponding second image pixel value matrix, and uses a preset image pixel value matrix to filter the noise of the first image pixel value matrix and the second image pixel value matrix, and generates and transmits a first processed image information and a second processed image information to the recognition unit to recognize the first processed image information to generate the corresponding first partial images, and recognize the second processed image information to generate the corresponding second partial images. According to the customized stereoscopic image production system described in claim 6, the computing processing unit uses the posture angle information and the posture part information to correspond to the first part images and the second part images, selects a plurality of corresponding image components, and reassembles the image components to generate the corresponding stereoscopic image model. The customized stereoscopic image production system according to claim 6 includes a stereoscopic printing device connected to the signal of the computing processing unit, receiving and generating a corresponding stereoscopic object according to the stereoscopic image model. According to the customized stereoscopic image production system described in claim 6, the capture unit is an ultrasonic scanner.

Images