CN113591961A - Minimally invasive medical camera image identification method based on neural network - Google Patents
Minimally invasive medical camera image identification method based on neural network Download PDFInfo
- Publication number
- CN113591961A CN113591961A CN202110831211.XA CN202110831211A CN113591961A CN 113591961 A CN113591961 A CN 113591961A CN 202110831211 A CN202110831211 A CN 202110831211A CN 113591961 A CN113591961 A CN 113591961A
- Authority
- CN
- China
- Prior art keywords
- neural network
- image
- minimally invasive
- method based
- invasive medical
- 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
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 74
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000012545 processing Methods 0.000 claims abstract description 11
- 238000012549 training Methods 0.000 claims abstract description 6
- 238000006243 chemical reaction Methods 0.000 claims abstract description 4
- 238000001914 filtration Methods 0.000 claims abstract description 4
- 210000005036 nerve Anatomy 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 238000002324 minimally invasive surgery Methods 0.000 abstract description 2
- 238000003909 pattern recognition Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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; 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
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a minimally invasive medical camera image identification method based on a neural network, which comprises the following steps: s1: firstly, acquiring target image information to be identified by using a CMOS camera; s2: then inputting the images in the sample library into a computer for conversion to obtain digital images, and then carrying out digital filtering processing to filter out some unnecessary information; s3: and then inputting the digital information of the sample image into a designed neural network for training and learning to generate an image recognition neural network system. According to the minimally invasive medical camera image identification method based on the neural network, the neural network composition can be quickly analyzed through the camera image identification method, and the neural network analysis process is saved; the electronic neural network simulated by the new assembly detects the neural network, is convenient to analyze the problems of the neural network and is beneficial to the implementation of minimally invasive surgery.
Description
Technical Field
The invention relates to the technical field of neural network medical treatment, in particular to a minimally invasive medical camera image identification method based on a neural network.
Background
In the minimally invasive research of the neural network, the identification of the neural network has great significance in the medical field, the neural network can be quickly identified conveniently, and the quick positioning of the neural problem is realized. For image processing and identification, mainly processes of converting or transforming an image and the like are performed to ensure that the image is effectively identified, the image mainly comprises two-dimensional space information of the image, the image processing mainly performs information impedance matching and appropriately adjusts amplitude, and the content is digitized and the like. The image recognition method used in the past is a correlation method, a moment invariant method and a projection method, and the research is more and more deep in the artificial intelligence theory at present, and particularly in the rapid development of computer technology, the method is more widely applied to neural network image recognition.
In the treatment process of the neural network, the components of the neural network are complex, and the neural network is analyzed slowly, so that the image recognition method of the minimally invasive medical camera based on the neural network is provided for solving the problems.
Disclosure of Invention
The invention aims to provide a minimally invasive medical camera image identification method based on a neural network, and aims to solve the problems that components of the neural network are complex and the analysis of the neural network is slow in the treatment process of the neural network provided by the background technology.
In order to achieve the purpose, the invention provides the following technical scheme: a minimally invasive medical camera image recognition method based on a neural network comprises the following steps:
s1: firstly, acquiring target image information to be identified by using a CMOS camera;
s2: then inputting the images in the sample library into a computer for conversion to obtain digital images, and then carrying out digital filtering processing to filter out some unnecessary information;
s3: and then inputting the digital information of the sample image into a designed neural network for training and learning to generate an image recognition neural network system.
Preferably, in step S1, the neural network picture is scanned by a CMOS camera, and the camera acquires the neural network picture.
Preferably, in step S1, during the image capturing process, the CMOS camera changes its orientation, and the captured target has distorted image information, which needs to be comprehensive, and the various distorted image information of the target constitutes a sample library for target identification.
The method has the advantages that the training speed of the neural network is improved, the reliability of the network is improved, the sensitivity of the network to local details of an error curved surface is reduced, the network is effectively restrained from being locally minimum, and the accuracy of target identification is improved.
Preferably, in step S1, the method of recognizing is to divide a feature or a group of features of the image into a plurality of fuzzy variables according to a certain blurring rule, make each fuzzy variable express a part of the features of the original feature, and then use the new fuzzy feature to replace the original feature for performing the fuzzy recognition.
Fuzzy pattern recognition is in essence a method or idea of introducing fuzzy logic into the pattern recognition problem. Fuzzy pattern recognition has been well applied in statistical pattern recognition, and the application of fuzzy theory in image recognition system is mainly to utilize fuzzy theory to fuzzify image features and to classify fuzzy features according to a certain fuzzification rule to divide one feature or a group of features of an image into a plurality of fuzzy variables, so that each fuzzy variable can express a part of the characteristics of the original feature, and then the original features are replaced by the new fuzzy features to carry out pattern recognition. Fuzzy classification is actually to divide the sample space into thousands of subsets, and these subsets are replaced by the concept of fuzzy subsets, so as to obtain fuzzy classification results, i.e. the fuzzification of the classification results. An advantage of a sample in fuzzy classification no longer belonging to a particular class but to a class with a different degree of confidence is:
(1) uncertainty in the classification process can be reflected in the classification result, and a user makes a decision according to the credibility;
(2) this can provide classification information for lower classes if multi-class classification is used.
Preferably, in step S2, the processor performs arithmetic processing on the information converted from the image.
Preferably, in step S2, the digitized information is identified and classified to obtain an accurate neural network route name, and the neural network route name is reattached to the original image and annotated to obtain a neural network diagram image.
Preferably, in step S3, the operation process is performed on the neural network with the identified nerves, and the connectivity of the neural network is detected, so as to obtain the neural network system generated by the image.
Compared with the prior art, the invention has the beneficial effects that: the minimally invasive medical camera image identification method based on the neural network comprises the following steps:
1. the neural network composition can be quickly analyzed by a camera image identification method, so that the neural network analysis process is saved;
2. the electronic neural network simulated by the new assembly detects the neural network, is convenient to analyze the problems of the neural network and is beneficial to the implementation of minimally invasive surgery.
Detailed Description
The technical solutions in the embodiments of the present invention are clearly and completely described, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention provides a technical scheme that: a minimally invasive medical camera image identification method based on a neural network is characterized in that: the minimally invasive medical camera image identification method based on the neural network comprises the following steps:
s1: firstly, acquiring target image information to be identified by using a CMOS camera;
s2: then inputting the images in the sample library into a computer for conversion to obtain digital images, and then carrying out digital filtering processing to filter out some unnecessary information;
s3: and then inputting the digital information of the sample image into a designed neural network for training and learning to generate an image recognition neural network system.
Further, in step S1, the neural network picture is scanned by using a CMOS camera, and the camera acquires the neural network picture.
Further, in step S1, in the image capturing process, the CMOS camera changes the orientation, and the captured target has distorted image information, which needs to be comprehensive, and the various distorted image information of the target constitutes a sample library for target identification.
The method has the advantages that the training speed of the neural network is improved, the reliability of the network is improved, the sensitivity of the network to local details of an error curved surface is reduced, the network is effectively restrained from being locally minimum, and the accuracy of target identification is improved.
Further, in step S1, a feature or a group of features of the image is divided into a plurality of fuzzy variables according to a certain blurring rule, each fuzzy variable can express a part of the features of the original feature, and then the new fuzzy feature is used to replace the original feature for performing the fuzzy recognition.
Fuzzy pattern recognition is in essence a method or idea of introducing fuzzy logic into the pattern recognition problem. Fuzzy pattern recognition has been well applied in statistical pattern recognition, and the application of fuzzy theory in image recognition system is mainly to utilize fuzzy theory to fuzzify image features and to classify fuzzy features according to a certain fuzzification rule to divide one feature or a group of features of an image into a plurality of fuzzy variables, so that each fuzzy variable can express a part of the characteristics of the original feature, and then the original features are replaced by the new fuzzy features to carry out pattern recognition. Fuzzy classification is actually to divide the sample space into thousands of subsets, and these subsets are replaced by the concept of fuzzy subsets, so as to obtain fuzzy classification results, i.e. the fuzzification of the classification results. An advantage of a sample in fuzzy classification no longer belonging to a particular class but to a class with a different degree of confidence is:
(1) uncertainty in the classification process can be reflected in the classification result, and a user makes a decision according to the credibility;
(2) this can provide classification information for lower classes if multi-class classification is used.
Further, in step S2, the processor performs arithmetic processing on the information converted from the image.
Further, in step S2, the digitized information is identified and classified to obtain an accurate neural network route name, and the neural network route name is reattached to the original image and annotated to obtain a neural network diagram image.
Further, in step S3, the operation processing is performed on the neural network with the identified nerves, and the connectivity of the neural network is detected, so as to obtain a neural network system generated from the image.
Although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that various changes in the embodiments and/or modifications of the invention can be made, and equivalents and modifications of some features of the invention can be made without departing from the spirit and scope of the invention.
Claims (7)
1. A minimally invasive medical camera image identification method based on a neural network is characterized in that: the minimally invasive medical camera image identification method based on the neural network comprises the following steps:
s1: firstly, acquiring target image information to be identified by using a CMOS camera;
s2: then inputting the images in the sample library into a computer for conversion to obtain digital images, and then carrying out digital filtering processing to filter out some unnecessary information;
s3: and then inputting the digital information of the sample image into a designed neural network for training and learning to generate an image recognition neural network system.
2. The minimally invasive medical camera image recognition method based on the neural network as claimed in claim 1, wherein: in step S1, the neural network picture is scanned by a CMOS camera, and the neural network picture is acquired by the camera.
3. The minimally invasive medical camera image recognition method based on the neural network as claimed in claim 1, wherein: in step S1, in the image collecting process, the CMOS camera changes the orientation, the collected targets have distorted image information, the distorted target information needs to be comprehensive, and the various distorted image information of the targets form a sample library for target identification.
4. The minimally invasive medical camera image recognition method based on the neural network as claimed in claim 1, wherein: in step S1, the identification method is to divide a feature or a group of features of the image into a plurality of fuzzy variables according to a certain fuzzy rule, so that each fuzzy variable can express a part of the features of the original feature, and then replace the original feature with the new fuzzy feature to perform fuzzy identification.
5. The minimally invasive medical camera image recognition method based on the neural network as claimed in claim 1, wherein: in step S2, the processor performs arithmetic processing on the information converted from the image.
6. The minimally invasive medical camera image recognition method based on the neural network as claimed in claim 1, wherein: in step S2, the digitized information is identified and classified to obtain an accurate neural network route name, and the neural network route name is reattached to the original image and annotated to obtain a neural network schematic image.
7. The minimally invasive medical camera image recognition method based on the neural network as claimed in claim 1, wherein: in step S3, the operation processing is performed on the neural network whose nerves have been identified, and the connectivity of the neural network is detected, so as to obtain a neural network system generated from the image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110831211.XA CN113591961A (en) | 2021-07-22 | 2021-07-22 | Minimally invasive medical camera image identification method based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110831211.XA CN113591961A (en) | 2021-07-22 | 2021-07-22 | Minimally invasive medical camera image identification method based on neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113591961A true CN113591961A (en) | 2021-11-02 |
Family
ID=78249061
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110831211.XA Pending CN113591961A (en) | 2021-07-22 | 2021-07-22 | Minimally invasive medical camera image identification method based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113591961A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109310306A (en) * | 2016-06-28 | 2019-02-05 | 索尼公司 | Image processing apparatus, image processing method and medical imaging system |
| CN109815945A (en) * | 2019-04-01 | 2019-05-28 | 上海徒数科技有限公司 | A kind of respiratory tract inspection result interpreting system and method based on image recognition |
| CN112348125A (en) * | 2021-01-06 | 2021-02-09 | 安翰科技(武汉)股份有限公司 | Capsule endoscope image identification method, equipment and medium based on deep learning |
| WO2021043193A1 (en) * | 2019-09-04 | 2021-03-11 | 华为技术有限公司 | Neural network structure search method and image processing method and device |
-
2021
- 2021-07-22 CN CN202110831211.XA patent/CN113591961A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109310306A (en) * | 2016-06-28 | 2019-02-05 | 索尼公司 | Image processing apparatus, image processing method and medical imaging system |
| CN109815945A (en) * | 2019-04-01 | 2019-05-28 | 上海徒数科技有限公司 | A kind of respiratory tract inspection result interpreting system and method based on image recognition |
| WO2021043193A1 (en) * | 2019-09-04 | 2021-03-11 | 华为技术有限公司 | Neural network structure search method and image processing method and device |
| CN112348125A (en) * | 2021-01-06 | 2021-02-09 | 安翰科技(武汉)股份有限公司 | Capsule endoscope image identification method, equipment and medium based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 雷建云，张勇，李海凤: "网络信息安全理论与技术", vol. 978, 31 March 2009, 中国商务出版社, pages: 112 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Singla et al. | Automated latent fingerprint identification system: A review | |
| Das et al. | Convolutional neural network for finger-vein-based biometric identification | |
| Manickam et al. | Score level based latent fingerprint enhancement and matching using SIFT feature | |
| Chen et al. | Segmentation of fingerprint images using linear classifier | |
| Hasan et al. | Fingerprint image enhancement and recognition algorithms: a survey | |
| Feng et al. | Robust and efficient algorithms for separating latent overlapped fingerprints | |
| Ghosh et al. | Bangla handwritten character recognition using MobileNet V1 architecture | |
| Rahman et al. | Person identification using ear biometrics | |
| Zhu et al. | FingerGAN: A constrained fingerprint generation scheme for latent fingerprint enhancement | |
| Thongtawee et al. | A novel feature extraction for American sign language recognition using webcam | |
| Jennifer et al. | A neutrosophic set approach on chest X-rays for automatic lung infection detection | |
| Manickam et al. | Level 2 feature extraction for latent fingerprint enhancement and matching using type-2 intuitionistic fuzzy set | |
| Raghavendra et al. | Utilization of Spatial Filtering for Enhancement of Fingerprint Images | |
| Achban et al. | Wrist hand vein recognition using local line binary pattern (LLBP) | |
| CN107103289B (en) | Method and system for handwriting identification using handwriting outline features | |
| CN113591961A (en) | Minimally invasive medical camera image identification method based on neural network | |
| Abdelrahaman et al. | K-nearest neighbor classifier for signature verification system | |
| Stojanovic et al. | Latent overlapped fingerprint separation: a review | |
| Khan et al. | An offline signature recognition and verification system based on neural network | |
| Balangue et al. | Ear Recognition for Ear Biometrics Using Integrated Image Processing Techniques via Raspberry Pi | |
| Aydoğdu et al. | A study on liveness analysis for palmprint recognition system | |
| Kehinde et al. | Hand geometry recognition: an approach for closed and separated fingers | |
| Cappelli | Unveiling the power of simplicity: Two remarkably effective methods for fingerprint segmentation | |
| Mahmood et al. | A new hand gesture recognition system using artificial neural network | |
| Sahu et al. | Fingerprint identification system using tree based matching |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Li Zhiqiang Inventor after: The teeth are fragrant Inventor after: Du Baoxue Inventor before: The teeth are fragrant Inventor before: Du Baoxue Inventor before: Zhou Zuoguang |
|
| CB03 | Change of inventor or designer information | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211102 |
|
| RJ01 | Rejection of invention patent application after publication |