CN111062961A - Contact lens edge defect detection method based on deep learning - Google Patents
Contact lens edge defect detection method based on deep learning Download PDFInfo
- Publication number
- CN111062961A CN111062961A CN201911402408.0A CN201911402408A CN111062961A CN 111062961 A CN111062961 A CN 111062961A CN 201911402408 A CN201911402408 A CN 201911402408A CN 111062961 A CN111062961 A CN 111062961A
- Authority
- CN
- China
- Prior art keywords
- image
- contact lens
- initial
- detection model
- defect
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 130
- 238000001514 detection method Methods 0.000 title claims abstract description 113
- 238000013135 deep learning Methods 0.000 title description 10
- 238000012549 training Methods 0.000 claims abstract description 46
- 238000000034 method Methods 0.000 claims abstract description 45
- 238000007781 pre-processing Methods 0.000 claims abstract description 16
- 238000003384 imaging method Methods 0.000 claims description 8
- 238000013526 transfer learning Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 abstract description 8
- 238000012545 processing Methods 0.000 abstract description 3
- 238000012795 verification Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 239000011521 glass Substances 0.000 description 3
- 230000002950 deficient Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/958—Inspecting transparent materials or objects, e.g. windscreens
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/0008—Industrial image inspection checking presence/absence
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/958—Inspecting transparent materials or objects, e.g. windscreens
- G01N2021/9583—Lenses
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Biochemistry (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Quality & Reliability (AREA)
- Testing Of Optical Devices Or Fibers (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
一种隐形眼镜边缘缺陷检测模型、其建立方法和检测方法，该检测模型的建立方法包括获取隐形眼镜边缘缺陷样本图像；对获取的样本图像预处理，得到样本特征图像；对样本特征图像进行缺陷标注，汇总标记的样本特征图像建立训练集；对训练集进行模型训练并进行微调后得到初始缺陷检测模型；检测初始缺陷检测模型，得到初始正确率，当初始正确率小于预设正确率时，调整初始缺陷检测模型后再次检测，当初始正确率大于或等于预设正确率时，即得到最终的所述缺陷检测模型。本发明克服了人工检测主观意识强、效率低、传统图像处理检测繁琐等问题，本发明的模型适用于对隐形眼镜的快速在线检测。
A contact lens edge defect detection model, its establishment method and detection method. The establishment method of the detection model includes acquiring a sample image of a contact lens edge defect; preprocessing the acquired sample image to obtain a sample feature image; Label, summarize the labeled sample feature images to establish a training set; perform model training on the training set and fine-tune to obtain an initial defect detection model; detect the initial defect detection model to obtain the initial correct rate, when the initial correct rate is less than the preset correct rate, After adjusting the initial defect detection model, test again, and when the initial accuracy rate is greater than or equal to the preset accuracy rate, the final defect detection model is obtained. The invention overcomes the problems of strong subjective consciousness, low efficiency, and complicated traditional image processing detection in manual detection, and the model of the invention is suitable for rapid online detection of contact lenses.
Description
Technical Field
The invention relates to the field of optical lens defect detection, in particular to a contact lens edge defect detection method based on deep learning.
Background
As the demand for contact lenses has increased, quality inspection of contact lenses has become increasingly important. For defect detection of contact lenses, the existing methods are mostly based on manual detection, and have the following defects: 1. long time, low efficiency and large number of personnel; 2. the subjective consciousness of manual detection is strong, and fatigue can occur after long-time work, so that the problems of false detection and missed detection can be caused. The defect detection method based on the traditional image processing needs complex image preprocessing, and realizes fussy feature extraction through a pattern recognition technology.
The deep learning based on the machine vision becomes a popular technology for the current defect detection, and the deep learning not only can quickly carry out target detection and target positioning, but also has lower cost and high resolution. The invention provides that the robustness of the model is improved by continuously adjusting the detection parameters through deep learning, and the generalization capability of the model is improved through machine learning, so that the accuracy of defect detection is improved.
Disclosure of Invention
In view of the above, it is a primary objective of the present invention to provide a contact lens edge defect detection model, a method for creating the same, and a method for detecting the same, which are intended to at least partially solve at least one of the above-mentioned problems.
In order to achieve the above object, according to an aspect of the present invention, there is provided a method for establishing a contact lens edge defect detection model, including:
(1) acquiring a sample image of the edge defect of the contact lens;
(2) preprocessing the obtained sample image to obtain a sample characteristic image;
(3) carrying out defect labeling on the sample characteristic images, summarizing the labeled sample characteristic images and establishing a training set;
(4) performing model training on the training set and performing fine adjustment to obtain an initial defect detection model;
(5) and detecting the initial defect detection model to obtain an initial accuracy, adjusting the initial defect detection model and then detecting again when the initial accuracy is less than a preset accuracy, and obtaining a final defect detection model when the initial accuracy is more than or equal to the preset accuracy.
As another aspect of the invention, a contact lens edge defect detection model is also provided, which is obtained by adopting the establishing method.
As a further aspect of the invention, there is also provided a method for detecting edge defects of a contact lens, comprising:
acquiring a detection image of a contact lens to be detected;
preprocessing the detection image to obtain a sample characteristic image corresponding to the detection image;
inputting the sample characteristic image into the contact lens edge defect detection model, and outputting a detection result, namely completing the detection of the contact lens edge defect.
Based on the above technical solution, the contact lens edge defect detection model, the establishment method and the detection method thereof according to the present invention have at least one of the following advantages over the prior art:
1. the invention provides a contact lens defect detection method based on deep learning, which solves the problems of strong subjective consciousness, low efficiency, fussy traditional image processing and detection and the like of manual detection;
2. according to the invention, through deep learning to train the mark sample with defects and continuous self-learning adjustment of parameters, a robust model with extremely high accuracy can be obtained, and the model can accurately extract defect characteristics, so that the rapid online detection of the contact lenses is realized.
Drawings
FIG. 1 is a flow chart of a method for establishing contact lens edge defect detection according to an embodiment of the present invention;
FIG. 2 is a sample view of a defective contact lens in an embodiment of the invention;
FIG. 3 is a sample diagram of the defect test results in the embodiment of the present invention.
Detailed Description
In order that the objects, technical solutions and advantages of the present invention will become more apparent, the present invention will be further described in detail with reference to the accompanying drawings in conjunction with the following specific embodiments.
The invention provides a novel detection method in view of the defects of the existing defect detection method of the contact lens in the prior art. The method is a contact lens defect detection method based on deep learning.
The invention discloses a method for establishing a contact lens edge defect detection model, which comprises the following steps:
(1) acquiring a sample image of the edge defect of the contact lens;
(2) preprocessing the obtained sample image to obtain a sample characteristic image;
(3) carrying out defect labeling on the sample characteristic images, summarizing the labeled sample characteristic images and establishing a training set;
(4) performing model training on the training set and performing fine adjustment to obtain an initial defect detection model;
(5) and detecting the initial defect detection model to obtain an initial accuracy, adjusting the initial defect detection model and then detecting again when the initial accuracy is less than a preset accuracy, and obtaining a final defect detection model when the initial accuracy is more than or equal to the preset accuracy.
In some embodiments of the present invention, the method for detecting the initial accuracy in step (5) comprises: and inputting the detected sample image into an initial defect detection model, automatically identifying and outputting a defect mark by the initial defect detection model, comparing with a manual detection result, and counting to obtain the initial accuracy.
In some embodiments of the present invention, the fine tuning method in the initial detection model forming process in step (4) includes a transfer learning technique.
In some embodiments of the present invention, the method for preprocessing the sample image in step (2) includes filtering, image equalization and image enhancement.
In some embodiments of the present invention, the sample image in step (1) is an image captured based on a preset imaging method.
In some embodiments of the present invention, the specimen images in step (1) comprise at least 1200 contact lens images with defects.
In some embodiments of the present invention, the contact lens of step (1) is a damaged-edge contact lens.
The invention also discloses a contact lens edge defect detection model obtained by adopting the establishing method.
The invention also discloses a method for detecting the edge defect of the contact lens, which comprises the following steps:
acquiring a detection image of a contact lens to be detected;
preprocessing the detection image to obtain a sample characteristic image corresponding to the detection image;
inputting the sample characteristic image into the contact lens edge defect detection model, and outputting a detection result, namely completing the detection of the contact lens edge defect.
In one exemplary embodiment, the method for establishing a novel contact lens edge defect detection model of the present invention comprises:
acquiring a contact lens sample image, and establishing a defect sample database; the specimen images include at least 1200 contact lens images with different types of defects;
preprocessing the sample image to obtain a sample characteristic image corresponding to the sample;
marking the defect area of the sample characteristic image, summarizing the marked sample image and establishing a training set;
performing model training on the training set, performing deep learning, and adjusting a score threshold value based on defect recognition to obtain an initial defect detection model;
inputting the sample image into the initial defect detection model, and outputting to obtain a predicted defect image corresponding to the sample image; the defect prediction image is a sample image of a defect area marked by the defect detection model;
and collecting test samples of the contact lenses, and establishing a contact lens defect test set. And judging whether the defect identification achieves the detection purpose or not based on the accuracy of the training model in the detection result of the test sample set. The accuracy rate set in this example was 95%.
And carrying out result statistics on the predicted defect image, and continuously adjusting a model according to a result to finally obtain the contact lens edge defect detection model.
The sample image is an image shot based on a preset imaging method;
wherein the defective contact lens is a damaged-edge contact lens.
The method comprises the following steps of carrying out model training on the training set, carrying out fine tuning by using a transfer learning technology, and outputting to obtain an initial defect detection model, and specifically comprises the following steps:
the method comprises the steps of carrying out initial training on a training set by utilizing a neural network to obtain a defect region of an initial prediction defect image, comparing the defect region of the initial prediction defect image with a defect region marked in a sample image, continuously updating training through comparison, and carrying out fine tuning by utilizing a transfer learning technology, wherein the fine tuning standard is a sub-training process and a verification set, when training, the network can use one part of input picture data as the training set, one part of the input picture data is divided into the verification set, parameters of training set training can be compared with the verification set, and iterative updating is carried out according to the comparison. And after the training is finished, fine adjustment is carried out again to form an initial defect detection model.
Wherein, the statistics of the predicted defect image result determines the accuracy of the defect detection model in the training process;
and when the accuracy is smaller than the preset accuracy, re-changing the corresponding parameters and the model selection and executing the training process until the accuracy is not smaller than the preset accuracy.
The embodiment also discloses a novel method for detecting edge defects of contact lenses, which comprises the following steps:
acquiring a detection image of glasses to be detected;
preprocessing the detection image to obtain a sample characteristic image corresponding to the detection image;
inputting the sample characteristic image into a defect detection model obtained by training according to the defect detection model training method, and outputting a predicted image of the detection image;
and determining the defects of the glasses to be detected based on the predicted image, judging whether the glasses to be detected contain the defects, comparing the defects with an artificial visual inspection result, and finishing the detection of the edge defects of the contact lenses.
In another exemplary embodiment, the method for detecting edge defects of a contact lens of the present invention comprises:
step A: acquiring a contact lens sample image under a specific imaging mode;
and B: preprocessing the sample image to obtain a sample characteristic image corresponding to the sample;
and C: marking the defect area of the sample characteristic image, summarizing the marked sample image and establishing a training set;
step D: performing model training on the training set, performing fine adjustment by using a transfer learning technology, and outputting to obtain an initial defect detection model;
step E: and D, inputting the sample image into an initial defect detection model, outputting to obtain a predicted defect image corresponding to the sample image, carrying out result statistics on the predicted defect image, continuously adjusting the model according to the result, and when the accuracy of the predicted defect image is lower than 95%, carrying out the step D again to adjust the model. And when the accuracy is greater than or equal to 95%, obtaining the final contact lens edge defect detection model.
Step F: and E, inputting the detection image of the contact lens to be detected into the contact lens edge defect detection model obtained in the step E for detection, namely completing the detection of the contact lens edge defect.
Wherein, in the imaging mode of the step A, at least one of the following methods is included: transmission, reflection and refraction, even in an imaging mode, change the brightness of a light source, and acquire images on the reverse surface and the front surface of the contact lens.
Wherein, the image preprocessing in the step B comprises: filtering, image equalization and image enhancement.
Wherein the defect area in step C is an edge damage area of the contact lens.
In the transfer learning technology, the initialization model is trained secondarily by using Faster R-CNN.
The technical solution of the present invention is further illustrated by the following specific embodiments in conjunction with the accompanying drawings. It should be noted that the following specific examples are given by way of illustration only and the scope of the present invention is not limited thereto.
The method for establishing the contact lens edge defect detection model of the present embodiment is shown in fig. 1, and specifically includes the following steps:
step A: selecting a specific imaging mode for one or more defects of the contact lens to make the defects prominent, and acquiring images by using an industrial CCD camera;
and B: preprocessing the acquired sample image to obtain a sample characteristic image corresponding to the sample;
and C: marking the defect area of the sample characteristic image, summarizing the marked sample image and establishing a training set;
step D: performing model training on the training set, performing fine adjustment by using a transfer learning technology, and outputting to obtain an initial defect detection model;
step E: inputting the sample image into an initial defect detection model, outputting to obtain a predicted defect image corresponding to the sample image, carrying out result statistics on the predicted defect image, and continuously adjusting the model according to the result. And when the accuracy of the predicted defect image is lower than 95%, the step D is carried out again to adjust the model. And when the accuracy is greater than or equal to 95%, obtaining the final contact lens edge defect detection model.
And E, inputting the detection image of the contact lens to be detected into the final contact lens edge defect detection model obtained in the step E for detection, namely completing the detection of the contact lens edge defect.
Specifically, in step a, the resolution of the acquired image may be from 1280 × 1024 to 2048 × 3072, and the imaging system acquires 1200 pictures, as shown in fig. 2, with clear boundaries and obvious defects.
And B, preprocessing the image in the step B, including filtering, image equalization and image enhancement, dividing the acquired image into a training set and a verification set and a test set according to the proportion of 8: 1, and carrying out defect marking on the training set.
And C, model training is carried out on 960 training set pictures and 120 verification set pictures, and a proper model frame and a convolution neural network are selected, wherein the model frame is one of fast R-CNN, Rception, ResNet and VGG. Setting an initial learning _ rate to be 0.01, iters _ num to be 100000, batch _ size to be 1, train _ size to be 120, setting a confidence parameter to be 0.8, calculating the error between a training set and a verification set by using a random gradient descent method, modifying a weight parameter according to the error, and fine-tuning by using fast R-CNN after the training is finished to obtain an initial training model so as to ensure the robustness and accuracy of the generated model. The fine tuning standard is that the sub-training process is compared with the verification set, and after the training is finished, the fine tuning is carried out to form an initial defect detection model
And for the generated model, performing final verification by using 120 test sets to perform result statistics on the predicted defect image, wherein the image prediction result is shown in fig. 3, the position of the defect is marked, the type of the defect and the defect are marked, and for the 120 test sets, when the accuracy is lower than 95%, selecting a neural network and setting various hyper-parameters, and retraining the model to obtain an expected model (namely the contact lens edge defect detection model).
The specific embodiments described herein are merely illustrative of the principles of the invention. Various modifications or additions may be made to the described embodiments or alternatives in a similar manner by those skilled in the art without departing from the principles of the present invention or exceeding the scope of the invention as defined in the appending claims.
Although the present invention is directed to contact lens edge defects, it does not exclude the possibility of using other terms, nor does it exclude other defect types.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention and are not intended to limit the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. A method for establishing a contact lens edge defect detection model comprises the following steps:
(1) acquiring a sample image of the edge defect of the contact lens;
(2) preprocessing the obtained sample image to obtain a sample characteristic image;
(3) carrying out defect labeling on the sample characteristic images, summarizing the labeled sample characteristic images and establishing a training set;
(4) performing model training on the training set and performing fine adjustment to obtain an initial defect detection model;
(5) and detecting the initial defect detection model to obtain an initial accuracy, adjusting the initial defect detection model and then detecting again when the initial accuracy is less than a preset accuracy, and obtaining a final defect detection model when the initial accuracy is more than or equal to the preset accuracy.
2. The method of establishing according to claim 1,
the method for detecting the initial accuracy in the step (5) comprises the following steps: and inputting the detected sample image into an initial defect detection model, automatically identifying and outputting a defect mark by the initial defect detection model, comparing with a manual detection result, and counting to obtain the initial accuracy.
3. The method of establishing according to claim 1,
and (4) the fine tuning method in the initial detection model forming process is a transfer learning method.
4. The method of establishing according to claim 1,
the sample image preprocessing method in the step (2) comprises filtering, image equalization and image enhancement.
5. The method of establishing according to claim 1,
the sample image in the step (1) is an image obtained by shooting based on a preset imaging method.
6. The method of establishing according to claim 1,
the specimen images in step (1) comprise at least 1200 contact lens images with defects.
7. The method of establishing according to claim 1,
the contact lens in the step (1) is a contact lens with damaged edge.
8. A contact lens edge defect detection model obtained by the method of any one of claims 1 to 7.
9. A method for detecting edge defects of a contact lens, comprising:
acquiring a detection image of a contact lens to be detected;
preprocessing the detection image to obtain a sample characteristic image corresponding to the detection image;
inputting the sample feature image into the contact lens edge defect detection model according to claim 8, and outputting the detection result, i.e. completing the detection of the contact lens edge defect.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911402408.0A CN111062961A (en) | 2019-12-30 | 2019-12-30 | Contact lens edge defect detection method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911402408.0A CN111062961A (en) | 2019-12-30 | 2019-12-30 | Contact lens edge defect detection method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111062961A true CN111062961A (en) | 2020-04-24 |
Family
ID=70305120
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911402408.0A Pending CN111062961A (en) | 2019-12-30 | 2019-12-30 | Contact lens edge defect detection method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111062961A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112529876A (en) * | 2020-12-15 | 2021-03-19 | 天津大学 | Method for detecting edge defects of contact lenses |
| CN114708266A (en) * | 2022-06-07 | 2022-07-05 | 青岛通产智能科技股份有限公司 | Tool, method and device for detecting card defects and medium |
| CN114742752A (en) * | 2022-03-02 | 2022-07-12 | 平方和(北京)科技有限公司 | Method and device for detecting printing quality of contact lens casting mold |
| US20230033187A1 (en) * | 2021-07-28 | 2023-02-02 | Coopervision International Limited | Systems and methods for acquiring and inspecting lens images of ophthalmic lenses |
| CN116990450A (en) * | 2023-07-18 | 2023-11-03 | 欧几里德(苏州)医疗科技有限公司 | Defect detection method and system for cornea shaping mirror |
| US20230400714A1 (en) * | 2022-06-08 | 2023-12-14 | Johnson & Johnson Vision Care, Inc. | Methods for quality control of contact lenses |
| EP4343302A1 (en) * | 2022-09-26 | 2024-03-27 | Essilor International | Computer-implemented method for determining a localization of at least one defect in a tested ophthalmic lens |
| CN117849072A (en) * | 2024-01-15 | 2024-04-09 | 平方和(北京)科技有限公司 | A contact lens casting mold printing defect detection method and system |
| CN119125008A (en) * | 2024-11-13 | 2024-12-13 | 吉林瑞尔康光学科技有限公司 | An artificial intelligence-based method for detecting defects in contact lens production |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE4124003A1 (en) * | 1991-07-19 | 1993-01-21 | Bodenseewerk Geraetetech | Optical contact lens testing device |
| JP2001523815A (en) * | 1997-11-14 | 2001-11-27 | ウェズリー ジェッセン コーポレイション | Automatic lens inspection system |
| CN205139045U (en) * | 2015-10-09 | 2016-04-06 | 南京爱丁堡环保科技有限公司 | Automated inspection equipment of contact lens defect |
| CN107341792A (en) * | 2017-06-21 | 2017-11-10 | 苏州卡睿知光电科技有限公司 | A kind of method and device for establishing the model for differentiating eyeglass defect type |
| CN109242829A (en) * | 2018-08-16 | 2019-01-18 | 惠州学院 | Liquid crystal display defect inspection method, system and device based on small sample deep learning |
-
2019
- 2019-12-30 CN CN201911402408.0A patent/CN111062961A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE4124003A1 (en) * | 1991-07-19 | 1993-01-21 | Bodenseewerk Geraetetech | Optical contact lens testing device |
| JP2001523815A (en) * | 1997-11-14 | 2001-11-27 | ウェズリー ジェッセン コーポレイション | Automatic lens inspection system |
| CN205139045U (en) * | 2015-10-09 | 2016-04-06 | 南京爱丁堡环保科技有限公司 | Automated inspection equipment of contact lens defect |
| CN107341792A (en) * | 2017-06-21 | 2017-11-10 | 苏州卡睿知光电科技有限公司 | A kind of method and device for establishing the model for differentiating eyeglass defect type |
| CN109242829A (en) * | 2018-08-16 | 2019-01-18 | 惠州学院 | Liquid crystal display defect inspection method, system and device based on small sample deep learning |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112529876B (en) * | 2020-12-15 | 2023-03-14 | 天津大学 | Method for detecting edge defects of contact lenses |
| CN112529876A (en) * | 2020-12-15 | 2021-03-19 | 天津大学 | Method for detecting edge defects of contact lenses |
| AU2022318275B2 (en) * | 2021-07-28 | 2024-04-18 | Coopervision International Limited | Acquiring and inspecting images of ophthalmic lenses |
| WO2023007110A1 (en) * | 2021-07-28 | 2023-02-02 | Coopervision International Limited | Acquiring and inspecting images of ophthalmic lenses |
| JP7602093B2 (en) | 2021-07-28 | 2024-12-17 | クーパーヴィジョン インターナショナル リミテッド | Acquiring and inspecting ophthalmic lenses |
| US20230033187A1 (en) * | 2021-07-28 | 2023-02-02 | Coopervision International Limited | Systems and methods for acquiring and inspecting lens images of ophthalmic lenses |
| US12051191B2 (en) * | 2021-07-28 | 2024-07-30 | Coopervision International Limited | Systems and methods for acquiring and inspecting lens images of ophthalmic lenses |
| KR20240014575A (en) * | 2021-07-28 | 2024-02-01 | 쿠퍼비젼 인터내셔널 리미티드 | Acquisition and examination of images of ophthalmic lenses |
| KR102676508B1 (en) | 2021-07-28 | 2024-06-24 | 쿠퍼비젼 인터내셔널 리미티드 | Acquisition and examination of images of ophthalmic lenses |
| CN114742752A (en) * | 2022-03-02 | 2022-07-12 | 平方和(北京)科技有限公司 | Method and device for detecting printing quality of contact lens casting mold |
| CN114708266A (en) * | 2022-06-07 | 2022-07-05 | 青岛通产智能科技股份有限公司 | Tool, method and device for detecting card defects and medium |
| US20230400714A1 (en) * | 2022-06-08 | 2023-12-14 | Johnson & Johnson Vision Care, Inc. | Methods for quality control of contact lenses |
| WO2023238084A1 (en) * | 2022-06-08 | 2023-12-14 | Johnson & Johnson Vision Care, Inc. | Methods for quality control of contact lenses |
| EP4343302A1 (en) * | 2022-09-26 | 2024-03-27 | Essilor International | Computer-implemented method for determining a localization of at least one defect in a tested ophthalmic lens |
| CN116990450B (en) * | 2023-07-18 | 2024-04-26 | 欧几里德(苏州)医疗科技有限公司 | Defect detection method and system for cornea shaping mirror |
| CN116990450A (en) * | 2023-07-18 | 2023-11-03 | 欧几里德(苏州)医疗科技有限公司 | Defect detection method and system for cornea shaping mirror |
| CN117849072A (en) * | 2024-01-15 | 2024-04-09 | 平方和(北京)科技有限公司 | A contact lens casting mold printing defect detection method and system |
| CN117849072B (en) * | 2024-01-15 | 2024-11-15 | 平方和(北京)科技有限公司 | Contact lens casting mold printing defect detection method and system |
| CN119125008A (en) * | 2024-11-13 | 2024-12-13 | 吉林瑞尔康光学科技有限公司 | An artificial intelligence-based method for detecting defects in contact lens production |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111062961A (en) | Contact lens edge defect detection method based on deep learning | |
| CN111325713B (en) | Neural network-based wood defect detection method, system and storage medium | |
| CN111160301B (en) | Tunnel disease target intelligent identification and extraction method based on machine vision | |
| CN111814850A (en) | Defect detection model training method, defect detection method and related device | |
| CN113706490B (en) | Wafer defect detection method | |
| CN113222938A (en) | Chip defect detection method and system and computer readable storage medium | |
| CN109946304A (en) | On-line detection system and detection method for surface defects of parts based on feature matching | |
| CN110889823A (en) | SiC defect detection method and system | |
| CN114926407A (en) | Steel surface defect detection system based on deep learning | |
| CN114627089B (en) | Defect identification method, defect identification device, computer equipment and computer readable storage medium | |
| CN110689524A (en) | No-reference online image definition evaluation method and system | |
| TWI747686B (en) | A defect detection method and a defect detection device | |
| CN117635603A (en) | System and method for detecting on-line quality of hollow sunshade product based on target detection | |
| CN114913086B (en) | Face image quality enhancement method based on generation countermeasure network | |
| CN111178405A (en) | A Similar Object Recognition Method Integrating Multiple Neural Networks | |
| CN117256514B (en) | Method for removing unqualified hatching eggs for sorting system | |
| CN109461155B (en) | Raw silk quality detection method | |
| CN106447045A (en) | Evaluation method for ADAS (Advanced Driver Assistant System) based on machine learning | |
| CN114286078B (en) | Camera module lens appearance inspection method and equipment | |
| CN113379689B (en) | Defect detection algorithm based on deep learning | |
| CN116245888A (en) | Steel defect image segmentation method based on the introduction of frequency domain deep learning network | |
| TWI777307B (en) | Method, computer program, and computer readable medium of using electroluminescence images to identify defect of solar cell based on deep learning technology | |
| CN116342502A (en) | A method of industrial vision detection based on deep learning | |
| Nguyen et al. | Detection of weak micro-scratches on aspherical lenses using a Gabor neural network and<? TeX\break?> transfer learning | |
| CN113267506A (en) | Wood board AI visual defect detection device, method, equipment and medium |
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 |