CN118196790B - Chromosome split phase image screening method, system, equipment and storage medium under low power lens - Google Patents
Chromosome split phase image screening method, system, equipment and storage medium under low power lens Download PDFInfo
- Publication number
- CN118196790B CN118196790B CN202410610271.2A CN202410610271A CN118196790B CN 118196790 B CN118196790 B CN 118196790B CN 202410610271 A CN202410610271 A CN 202410610271A CN 118196790 B CN118196790 B CN 118196790B
- Authority
- CN
- China
- Prior art keywords
- detected
- region
- connected domain
- information
- area
- 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
- 210000000349 chromosome Anatomy 0.000 title claims abstract description 61
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000012216 screening Methods 0.000 title claims abstract description 25
- 239000006185 dispersion Substances 0.000 claims abstract description 24
- 238000002372 labelling Methods 0.000 claims abstract description 12
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 12
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000001746 injection moulding Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000022131 cell cycle Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/698—Matching; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/695—Preprocessing, e.g. image segmentation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention provides a method, a system, equipment and a storage medium for screening chromosome splitting phase images under a low-power mirror, wherein the method comprises the steps of marking a region to be detected containing chromosome splitting phases on an original slide image by using a boundary box through a pre-trained boundary labeling model, scoring the region to be detected through the communicating domain dispersion and the communicating domain confidence of the region to be detected in the process of traversing the slide image, and accurately identifying and selecting the optimal splitting phase image through scoring and sorting.
Description
Technical Field
The invention relates to the field of chromosome screening, in particular to a method, a system, equipment and a storage medium for screening chromosome splitting phase images under a low-power mirror.
Background
Chromosome karyotyping is a complex process that typically requires multiple steps to obtain accurate results. In a karyotyping procedure, a full-slice scan under a low power mirror is typically performed first, which is intended to rapidly screen out chromosome images with appropriate split phases for subsequent high power mirror scanning and further analysis. However, under the low power mirror, the split phase image of the chromosome has the characteristics of smaller area and relatively dense, and the characteristics can lead to a certain error in chromosome identification, so that the split phase image is selected erroneously, further, the image obtained by scanning by the high power mirror is invalid, and time and resources are wasted.
In addition, not all split phases are suitable for karyotyping in images scanned under a low magnification. The morphology and structure of chromosomes may change due to different stages of the cell cycle, the biggest problem at present is that the chromosomes overlap each other, the distance between the chromosomes is too short, the splitting is extremely difficult to judge relative to image processing and analysts, and the overlapping positions are too many, so that the practicability of nuclear type analysis is very low; the method has the advantages that high requirements are put on the professional level and experience of analysts, the automation difficulty of the analysis process is high, the scanning flow is complex, and if the split phase shot under the low power mirror does not meet the analysis requirements of the analysts, the later stage is changed to the high power mirror shooting, so that time is wasted, and the time cost is increased.
Disclosure of Invention
In order to solve the problems, the invention provides a method, a system, equipment and a storage medium for screening chromosome splitting phase images under a low power lens.
The main content of the invention comprises:
a chromosome splitting phase image screening method under a low-power lens comprises the following steps:
Acquiring an original image; the original image is a slide image photographed by a low-power mirror and comprising a plurality of chromosome splitting phase images;
marking the original image to obtain a plurality of areas to be detected; the region to be detected comprises a marked boundary frame and a chromosome splitting phase image enclosed in the boundary frame;
respectively acquiring connected domain information in a single region to be detected; the single connected domain information comprises corresponding connected domain dispersion and connected domain confidence;
Calculating to obtain a scoring value of the region to be detected according to the connected domain dispersion degree and the connected domain confidence degree of the single region to be detected; the calculation formula is as follows:
Wherein, As the scoring value of the area to be detected,For the connected domain confidence of the region to be detected,For the connected domain dispersion of the region to be detected,、Is a weight coefficient;
And acquiring corresponding chromosome splitting phase images of the region to be detected according to the number required by high-power mirror shooting and the sequence from high to low of the scoring value of the region to be detected, and dividing the chromosome splitting phase images from the original images to form target candidate images.
Preferably, labeling the original image to obtain a plurality of areas to be detected, including:
and marking a corresponding boundary box for the chromosome splitting phase image on the original image by using a pre-trained boundary marking model.
Preferably, the acquiring the connected domain information in the single to-be-detected area includes:
performing binarization processing on the region to be detected to obtain a plurality of corresponding connected domain units;
respectively acquiring area information and perimeter information of the single connected domain unit;
Calculating first information of the region to be detected according to the area information and the perimeter information of all connected region units;
Calculating the connected domain dispersion of the region to be detected according to the first information and the direction information of the region to be detected;
the direction information of the area to be detected is the value of the second moment of the area to be detected.
Preferably, the first information of the region to be detected is calculated according to the area information and the perimeter information of all the connected domain units; comprising the following steps:
Calculating the perimeter of the connected domain unit Area of the connected domain unit；
The circumference information D of all connected domain units is calculated, and the calculation formula is as follows: wherein, the method comprises the steps of, wherein, The circumference of the ith connected domain unit is the circumference of the ith connected domain unit, and n is the number of the connected domain units in the to-be-detected area;
the area information S of all connected domain units is calculated, and the calculation formula is as follows: wherein, the method comprises the steps of, wherein, The area of the ith connected domain unit is the area, and n is the number of the connected domain units in the area to be detected;
calculating first information R of the region to be detected, wherein a calculation formula is as follows: 。
preferably, the direction information of the area to be detected is obtained by the following formula:
Wherein, Refers to the value of the second moment of the region to be detected,A pixel value representing a certain pixel point in the area to be detected,Representing the pixel average value of all pixel points in the area to be detected,Representing the number of pixels in the area to be detected.
Preferably, the dispersion of the area to be detected is obtained by the following formula:
Wherein, The dispersion of the region to be detected;， is a pre-configured weight coefficient.
The invention also provides a chromosome splitting phase image screening system under the low power lens, which comprises the following steps:
The acquisition module is used for acquiring slide images which are shot by the low-power mirror and contain a plurality of chromosome splitting phase images;
the marking module marks a boundary frame containing a chromosome splitting phase image in the slide image by using a boundary marking model, and divides the slide image into a plurality of areas to be detected;
The calculation module is used for acquiring the connected domain information in the region to be detected and calculating to obtain the corresponding score of the region to be detected according to the connected domain information;
The sequencing module is used for outputting chromosome split phase images of the corresponding region to be detected according to the number required by high-power mirror shooting and the sequence from high to low of the scoring value of the region to be detected;
The segmentation module is used for segmenting the selected region to be detected from the original image to form a target candidate image;
and the training module is used for training the boundary labeling model.
The invention also provides equipment, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize the chromosome splitting phase image screening method under the low power mirror.
The invention also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program realizes the chromosome splitting phase image screening method under the low power lens when being executed by a processor.
The invention has the beneficial effects that: the invention provides a method, a system, equipment and a storage medium for screening chromosome splitting phase images under a low-power mirror, wherein a boundary box is used for marking a region to be detected containing chromosome splitting phases on an original slide image through a pre-trained boundary labeling model, the region to be detected is scored in the process of traversing the slide image, the optimal splitting phase images can be accurately identified and selected through scoring and sorting, the requirement on the professional skills of operators is reduced, and the method can be used for quickly screening out chromosome splitting phase images meeting the requirement from a large number of low-power mirror images, so that the working efficiency is greatly improved, and the time and labor required by manual screening are reduced.
Drawings
Fig. 1 is a functional block diagram of the present invention.
Detailed Description
The technical scheme protected by the invention will be specifically described below with reference to the accompanying drawings.
Referring to fig. 1, the invention provides a system for screening a chromosome splitting phase image under a low power mirror, which specifically comprises an acquisition module, a marking module, a calculation module, a sequencing module and a segmentation module, wherein the acquisition module is used for acquiring a slide image which is shot by the low power mirror and contains a plurality of chromosome splitting phase images; that is, each slide shot by the low power mirror has a plurality of chromosome splitting phase images, and for convenience of description, a whole slide image is called an original image; according to the method, all split-phase images are not required to be firstly segmented from a slide and then screened, and a boundary box containing chromosome split-phase images is marked in the slide image by a marking module by using a boundary marking model so as to divide the slide image into a plurality of areas to be detected; the boundary labeling model is trained in advance, namely the system also comprises a training module, wherein the training module is used for training the used network model, for example, training the boundary labeling model, the image data set with the target can be collected for labeling, namely, the labeling comprises the boundary frame information of the target; the generalization capability and robustness of the model are improved by preprocessing the image, including size normalization, data enhancement, etc., and then training the selected model using the prepared dataset to learn how to accurately detect the location and bounding box of the object from the image.
The calculation module can acquire the connected domain information in the region to be detected, and calculate and obtain the corresponding score of the region to be detected according to the connected domain information; then, outputting chromosome split phase images of the corresponding areas to be detected according to the sequence from high to low of scoring values of the areas to be detected by the sequencing module according to the quantity required by high-power mirror shooting; and finally, dividing the selected region to be detected from the original image by a dividing module to form a target candidate image.
Specifically, the invention provides a chromosome splitting phase image screening method under a low-power lens, which comprises the following steps:
Firstly, marking a corresponding boundary box for a chromosome splitting phase image on an original image by using a pre-trained boundary marking model, namely dividing the chromosome splitting phase image in a slide image into a plurality of areas to be detected; the region to be detected comprises a marked boundary frame and a chromosome splitting phase image enclosed in the boundary frame; in other embodiments, the boundary labeling injection molding can be further trained, so that the boundary labeling injection molding can perform preliminary quality screening on the chromosome splitting phase images, for example, the chromosome quantity in the corresponding region to be detected is identified, the subsequent calculation analysis is not performed on the chromosome splitting phase images with the chromosome quantity being obviously smaller than a set value, and the boundary labeling model does not need to have a complicated classification and identification function, so that the region to be analyzed and calculated is rapidly divided.
Then, respectively acquiring the connected domain information in the single region to be detected; and respectively acquiring the connected domain dispersion degree and the connected domain confidence degree of all the areas to be detected.
Specifically, according to perimeter information and area information corresponding to a plurality of connected domains contained in a region to be detected, first information of the region to be detected is calculated; and then calculating the connected domain dispersion of the corresponding region to be detected by combining the connected domain dispersion with the second moment of the region to be detected, and finally scoring the connected domain confidence of the region to be detected by combining the connected domain confidence of the region to be detected. The score calculation formula is:
Wherein, As the scoring value of the area to be detected,For the connected domain confidence of the region to be detected,For the connected domain dispersion of the region to be detected,、For the weight coefficient, more specifically,、Is super-parameter, and can be normalized in the initial stage、All were set to 0.5, i.e., the degree of importance of connected domain confidence and connected domain dispersion in the final score was the same. Then, according to the scoring result, the method can also be manually adjusted、For adjustment, if the influence of the dispersion of the connected domain is desired to be increased, the height can be correspondingly adjustedTo be reduced。
Wherein, the connected domain confidence of the region to be detectedCan be calculated by yolov networks. The calculation of the connected domain dispersion of the to-be-detected area is obtained according to the first information and the direction information of the to-be-detected area, wherein the direction information of the to-be-detected area is the value of the second moment of the to-be-detected area, and the calculation is obtained by the following formula:
Wherein, Refers to the value of the second moment of the region to be detected,A pixel value representing a certain pixel point in the area to be detected,Representing the pixel average value of all pixel points in the area to be detected,Representing the number of pixels in the area to be detected.
The specific steps of respectively acquiring the connected domain information in the single region to be detected are as follows: firstly, binarizing a region to be detected to obtain a plurality of corresponding connected domain units; then, area information and perimeter information of the single connected domain unit are respectively obtained; then, calculating first information of the region to be detected according to the area information and the perimeter information of all connected domain units; in this embodiment, the perimeter of the connected domain unit is calculated firstArea of the connected domain unit; Then adding the areas of all connected domain units to obtain the area information S of the region to be detected, namelyAt the same time, adding the circumferences of all connected domain units to obtain the circumference information D of the region to be detected, namely; Wherein, For the perimeter of the i-th connected domain unit,The area of the ith connected domain unit is the area, and n is the number of the connected domain units in the area to be detected; and then useAnd obtaining the first information R of the region to be detected.
The dispersion of the region to be detected is obtained by the following formula:
Wherein, The dispersion of the region to be detected;， For pre-configured weight coefficients, in particular, 、Also superparameter, e.g. at initial stage, can be normalized、Setting the values to be the same, such as 0.5, and manually adjusting according to the final grading condition、Making adjustments, e.g. by enlargingTo increase the influence of the shape information of the region to be detected on the dispersion and reduceTo reduce the influence of the direction information on the dispersion.
And finally, according to the quantity required by high-power mirror shooting, acquiring corresponding chromosome splitting phase images of the region to be detected according to the sequence from high score value to low score value of the region to be detected, and dividing the chromosome splitting phase images from the original images to form target candidate images.
The invention also provides equipment, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize the chromosome splitting phase image screening method under the low power mirror.
The invention also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program realizes the chromosome splitting phase image screening method under the low power lens when being executed by a processor.
The foregoing description is only illustrative of the present invention and is not intended to limit the scope of the invention, and all equivalent structures or equivalent processes or direct or indirect application in other related technical fields are included in the scope of the present invention.
Claims (7)
1. The chromosome splitting phase image screening method under the low-power lens is characterized by comprising the following steps of:
Acquiring an original image; the original image is a slide image photographed by a low-power mirror and comprising a plurality of chromosome splitting phase images;
marking the original image to obtain a plurality of areas to be detected; the region to be detected comprises a marked boundary frame and a chromosome splitting phase image enclosed in the boundary frame;
respectively acquiring connected domain information in a single region to be detected; the single connected domain information comprises corresponding connected domain dispersion and connected domain confidence;
Calculating to obtain a scoring value of the region to be detected according to the connected domain dispersion degree and the connected domain confidence degree of the single region to be detected; the calculation formula is as follows:
，
Wherein, As the scoring value of the area to be detected,For the connected domain confidence of the region to be detected,For the connected domain dispersion of the region to be detected,、Is a weight coefficient;
According to the quantity required by high-power mirror shooting, obtaining corresponding chromosome splitting phase images of the region to be detected according to the sequence from high to low of scoring values of the region to be detected, and dividing the chromosome splitting phase images from an original image to form a target candidate image;
the method for respectively acquiring the connected domain information in the single to-be-detected area comprises the following steps:
performing binarization processing on the region to be detected to obtain a plurality of corresponding connected domain units;
respectively acquiring area information and perimeter information of the single connected domain unit;
Calculating first information of the region to be detected according to the area information and the perimeter information of all connected region units;
Calculating the connected domain dispersion of the region to be detected according to the first information and the direction information of the region to be detected;
The direction information of the region to be detected is the value of the second moment of the region to be detected;
calculating first information of the region to be detected according to the area information and the perimeter information of all connected region units; comprising the following steps:
Calculating the perimeter of the connected domain unit Area of the connected domain unit；
The circumference information D of all connected domain units is calculated, and the calculation formula is as follows: wherein, the method comprises the steps of, wherein, The circumference of the ith connected domain unit is the circumference of the ith connected domain unit, and n is the number of the connected domain units in the to-be-detected area;
the area information S of all connected domain units is calculated, and the calculation formula is as follows: wherein, the method comprises the steps of, wherein, The area of the ith connected domain unit is the area, and n is the number of the connected domain units in the area to be detected;
calculating first information R of the region to be detected, wherein a calculation formula is as follows: 。
2. the method for screening a chromosome splitting phase image under a low power mirror according to claim 1, wherein the method for marking an original image to obtain a plurality of regions to be detected comprises the steps of:
and marking a corresponding boundary box for the chromosome splitting phase image on the original image by using a pre-trained boundary marking model.
3. The method for screening a chromosome splitting phase image under a low power mirror according to claim 1, wherein the direction information of the region to be detected is obtained by the following formula:
，
Wherein, Refers to the value of the second moment of the region to be detected,A pixel value representing a certain pixel point in the area to be detected,Representing the pixel average value of all pixel points in the area to be detected,Representing the number of pixels in the area to be detected.
4. The method for screening the chromosome splitting phase image under the low-power mirror according to claim 1, wherein the connected domain dispersion of the region to be detected is obtained by the following formula:
，
Wherein, The connected domain dispersion of the region to be detected;， is a pre-configured weight coefficient.
5. A system using a low power microscopic chromosome splitting phase image screening method according to any of claims 1 to 4, comprising:
The acquisition module is used for acquiring slide images which are shot by the low-power mirror and contain a plurality of chromosome splitting phase images;
the marking module marks a boundary frame containing a chromosome splitting phase image in the slide image by using a boundary marking model, and divides the slide image into a plurality of areas to be detected;
The calculation module is used for acquiring the connected domain information in the region to be detected and calculating to obtain the corresponding score of the region to be detected according to the connected domain information;
The sequencing module is used for outputting chromosome split phase images of the corresponding region to be detected according to the number required by high-power mirror shooting and the sequence from high to low of the scoring value of the region to be detected;
The segmentation module is used for segmenting the selected region to be detected from the original image to form a target candidate image;
and the training module is used for training the boundary labeling model.
6. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a low power microscopic chromosome splitting phase image screening method according to any of claims 1 to 4.
7. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements a low power microscopic chromosome splitting phase image screening method according to any of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410610271.2A CN118196790B (en) | 2024-05-16 | 2024-05-16 | Chromosome split phase image screening method, system, equipment and storage medium under low power lens |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410610271.2A CN118196790B (en) | 2024-05-16 | 2024-05-16 | Chromosome split phase image screening method, system, equipment and storage medium under low power lens |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118196790A CN118196790A (en) | 2024-06-14 |
| CN118196790B true CN118196790B (en) | 2024-09-06 |
Family
ID=91403010
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410610271.2A Active CN118196790B (en) | 2024-05-16 | 2024-05-16 | Chromosome split phase image screening method, system, equipment and storage medium under low power lens |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118196790B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112037185A (en) * | 2020-08-21 | 2020-12-04 | 湖南自兴智慧医疗科技有限公司 | Chromosome split phase image screening method and device and terminal equipment |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110838113B (en) * | 2019-11-08 | 2022-03-04 | 南京大学金陵学院 | Method for detecting monofilament count and monofilament thickness consistency in multifilament synthesis |
| CN111325711A (en) * | 2020-01-16 | 2020-06-23 | 杭州德适生物科技有限公司 | Chromosome split-phase image quality evaluation method based on deep learning |
| CN111914818B (en) * | 2020-09-21 | 2024-05-24 | 北京林业大学 | A forest fire smoke root node detection method based on multi-frame discrete confidence |
| KR102630183B1 (en) * | 2021-12-21 | 2024-01-25 | 동아대학교 산학협력단 | Generating apparatus and method of image data for fire detection training, and learning apparatus and method using the same |
| KR102430071B1 (en) * | 2022-02-08 | 2022-08-05 | 국방과학연구소 | Image preprocessing method and apparatus based on labeling bounding box |
| CN114862877B (en) * | 2022-05-27 | 2024-03-22 | 四川大学华西医院 | Cell adhesion segmentation method and device based on confidence score |
-
2024
- 2024-05-16 CN CN202410610271.2A patent/CN118196790B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112037185A (en) * | 2020-08-21 | 2020-12-04 | 湖南自兴智慧医疗科技有限公司 | Chromosome split phase image screening method and device and terminal equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118196790A (en) | 2024-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111062885B (en) | Sign detection model training and sign detection method based on multi-stage transfer learning | |
| CN110543837A (en) | A Visible Light Airport Aircraft Detection Method Based on Potential Target Points | |
| CN111931751B (en) | Deep learning training method, target object identification method, system and storage medium | |
| CN108921057B (en) | Convolutional neural network-based prawn form measuring method, medium, terminal equipment and device | |
| CN113205511B (en) | Electronic component batch information detection method and system based on deep neural network | |
| CN116503622A (en) | Data acquisition and reading method based on computer vision image | |
| CN107305691A (en) | Foreground segmentation method and device based on images match | |
| CN112036384B (en) | Sperm head shape recognition method, device and equipment | |
| CN110827312A (en) | Learning method based on cooperative visual attention neural network | |
| CN117496532A (en) | Intelligent recognition tool based on 0CR | |
| CN107301421A (en) | The recognition methods of vehicle color and device | |
| CN117911409B (en) | Mobile phone screen bad line defect diagnosis method based on machine vision | |
| CN117850032B (en) | Optical lens adjusting method and system | |
| CN118196790B (en) | Chromosome split phase image screening method, system, equipment and storage medium under low power lens | |
| CN116737982B (en) | Intelligent screening management system for picture search results based on data analysis | |
| CN111260625B (en) | Automatic extraction method for offset printing large image detection area | |
| CN110276260B (en) | A product detection method based on depth camera | |
| CN118229629A (en) | Edge detection method, system, equipment and medium for large-size liquid crystal display | |
| CN115222652A (en) | Method for identifying, counting and centering end faces of bundled steel bars and memory thereof | |
| CN118506360B (en) | A method for recognizing dicentric chromosome images based on CNN convolutional neural network | |
| CN118918585B (en) | Automatic recognition method and system of pathological images based on deep learning | |
| CN112307908B (en) | Video semantic extraction method and device | |
| CN118864605B (en) | A target recognition and positioning system based on image segmentation and its terminal device | |
| CN118429780B (en) | Image processing method and device based on deep learning and computer equipment | |
| Bu et al. | Recognition of buildings on satellite images |
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 |