CN105389805A - Optical detection method of component shortage, wrong component and reverse component in plug-in operation - Google Patents
Optical detection method of component shortage, wrong component and reverse component in plug-in operation Download PDFInfo
- Publication number
- CN105389805A CN105389805A CN201510698911.0A CN201510698911A CN105389805A CN 105389805 A CN105389805 A CN 105389805A CN 201510698911 A CN201510698911 A CN 201510698911A CN 105389805 A CN105389805 A CN 105389805A
- Authority
- CN
- China
- Prior art keywords
- image
- analyzed
- pattern
- component
- value
- 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
Classifications
-
- 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
- G06T7/0006—Industrial image inspection using a design-rule based approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides an optical detection method of a component shortage, a wrong component and a reverse component in plug-in operation. The method comprises the steps of (1) using a camera to collect an image in the RGB format of a component in a plug-in unit, (2) carrying out image pre-processing to filter all information which can not display an image caused by the external factor interference in the imaging of the image or an unconcerned part in the image, and then converting the image in the RGB format into an image in the U8 format, (3) carrying out feature extraction analyzing of carrying out feature extraction on the RGB format image and the U8 format image of a concerned part left after the image pre-processing, using multiple image algorithms to orderly judge the shortage, wrong and reverse of the product component, and outputting a detection result, (4) presenting daily poor test ranking in the form of a report, and storing test data to facilitate the future tracking and query. According to online optical detection equipment, the technology of replacing human eyes by a camera is used, a target position is accurately and effectively judged through image processing algorithms, thus the judgment criterion is consistent, and the control of the appearance and quality of the product can be greatly improved.
Description
Technical field
The present invention relates to the field of optical detection in electronic information technology, be specifically related to a kind of for components and parts in part mate few part, wrong part and reverse optical detecting method, be applicable to the integrated circuit (IC) products that power circuit etc. has a large capacity and a wide range.
Background technology
For household electrical appliances auxiliary products after SMT, personnel's part mate has part often by wrong plug, and part is inserted less, part by anti-inserted to etc. situation, these problems need quality operating personnel to differentiate, and operating personnel easily produces visual fatigue by long visual identification, thus reduce judging nicety rate.
Summary of the invention
The invention provides components and parts in a kind of part mate few part, wrong part and reverse optical detecting method, online optical detection apparatus utilizes camera to replace human eye technology, by image processing algorithm, target location is carried out to the differentiation of precise and high efficiency, make discrimination standard consistance, significantly can promote the management and control to product appearance quality, reduce end user and differentiate.
For solving the problems of the technologies described above, the present invention adopts following technical scheme:
In part mate, components and parts few part, wrong part and a reverse optical detecting method, comprises the steps:
1) the rgb format image of components and parts in collected by camera plug-in unit is used;
2) Image semantic classification: to image when imaging because of extraneous factor interference cause imaging cannot show image full detail or to image in do not pay close attention to part and carry out filtering, then rgb format image is converted to U8 form;
3) feature-extraction analysis: carry out feature extraction respectively to the rgb format image of the concern part stayed after Image semantic classification and U8 format-pattern, adopts multiple image algorithm to judge few part of product components and parts, wrong part and oppositely successively, and output detections result;
4) with report form represent test every day defective products seniority among brothers and sisters, convenient storage is carried out to test data and follows the trail of inquiry in the future.
Further, the method for described Image semantic classification comprises following method:
Gradation conversion: gray level image U8 form is converted into RGB image, the color that filtering is abundant, only stays black in the white and black image to white middle transition pixel value;
Picture contrast, brightness regulation: by increasing integral image pixel value or reducing the brightness of image, by increasing the pixel value of the larger pixel of gray-scale value and reducing the less pixel value of gray-scale value, improve the contrast of light and shade object;
Filtering: the noise produced in shooting by gaussian filtering, mean filter and median filtering algorithm removal image;
Binaryzation: RGB image is converted to gray level image, by comparing to pixel each in image and setting value, is 1 for the value pixel transitions being greater than setting, otherwise is 0 converted image;
Image algorithm converts: to image carry out pixel add, subtract, with or logical relation computing change;
Particle analysis: the hollow sectors that in the fine particle of object edge or blank map picture, region to be analyzed is tiny is removed to the image converting gray scale to.
Further, in step 3) feature-extraction analysis, the image algorithm of described rgb format image comprises:
Color of image extraction and analysis: the light being reflected different colours by light-illuminating on object, object color is obtained by the GTG that the analysis of light extracted to often kind of light or the Cumulative sum that carries out GTG, differentiate whether the color of object meets the requirements by the color of getting, whether judgment object is because the wrong part of color or few part;
Whether color of image is in conjunction with local shape analysis: consistent with the object needed by extracting the profile combination judgment object CF excessively formed between grey decision-making and GTG reflected from object, determines wrong part, few part or reverse.
Further, in step 3) feature-extraction analysis, the image algorithm of described U8 format-pattern comprises:
Image Edge-Detection analyze: RGB image being converted to gray level image, determining the number of excessive point, line, surface by arranging value excessive between GTG, utilize number come judgment object whether reverse or lack part;
Whether geometric configuration detects is analyzed: change RGB image into gray level image, be combined to form obvious contour of object carry out rendering image shape by the transitional region between GTG, analyzes consistent with object under test profile, and whether judgment object lacks part or reverse;
Brightness detects to be analyzed: RGB image is converted to gray level image, asks for maximal value, minimum value and accumulated value to region to be analyzed grey decision-making in image, judge whether the luminous object in analyzed region exists;
Whether less contour detecting is analyzed: RGB image being converted to gray level image, filtering out unwanted pattern by arranging value excessive between GTG, retains the pattern needed, by pattern correspondance's judgment object part or wrong part;
Yardstick is measured to detect and is analyzed: go out by image and actual photographed areal calculation the actual physical size that single pixel occupies, area, the distance of pattern is calculated by the pixel value extracting region to be analyzed in image, utilize the distance of trying to achieve or area whether normal to the distance or area that judge provincial characteristics to be analyzed, whether judgment object is reverse, wrong part or few part;
Charactron reading Analysis: by zooming in or out the feature of seven segment code in the charactron of image region to be analyzed, distinguish different display pattern, according to the content definition of different pattern being differentiated to pattern displaying, judge charactron luminescence display whether normal or charactron whether less part by the content of definition;
Text region is analyzed: extract zone map to be analyzed in image, analyzing pattern names hereof with word or indications mode, name is carried out to similar pattern and extracts word or indications, by carrying out judgment object whether less part or wrong part to the word read or indications.
Further, when object is for motion or meeting consecutive variations state, described optical detecting method also comprises image characteristics extraction Avi videoing skill: by arranging Image Pretreatment Algorithm in advance, carries out the reservation of processed group synthetic video hereof to every two field picture of camera shooting.
From above technical scheme, the present invention is processed image by abundant algorithm, and programming is simple, and reserved external interface extensibility is strong, regulates camera at switching at runtime within the vision by program transformation camera parameter simultaneously; Realize multi-cam to take simultaneously, Synchronization Analysis is carried out to the image gathered, greatly shorten the test duration, can realize analyzing dynamic avi format video, avi format video is recorded by software itself, parameter filtering not focus again through setting in advance, thus realization realizes the analysis to frame of video to the extraction of video.
Embodiment
Below a kind of preferred implementation of the present invention is described in detail.
Can realize the few part of product components and parts, wrong part and reverse analysis by combination between Image semantic classification and the analysis of software built-in algorithms, represent with report form and test defective products seniority among brothers and sisters every day, convenient storage is carried out to test data and follows the trail of inquiry in the future, greatly reduce defective products rate of outflow, make product appearance quality obtain management and control.Below illustrating of image pre-processing method and image algorithm.
Image semantic classification be to image when imaging because of extraneous factor interference cause imaging cannot show image full detail or to image in do not pay close attention to part and carry out filtering, the image leaving concern part carries out feature extraction, the principle of process be feature according to specific needs to carry out pre-treatment step selection, main preprocess method is as follows:
Gradation conversion: be converted into gray level image to RGB image, the color that filtering is abundant, only stays black in the white and black image to white middle transition pixel value.Be applicable to object: needs are converted to U8 (without symbol 8 bit image image, i.e. greyish white image) image that form processes again, following algorithm depends on the process of U8 format-pattern: technique of image edge detection, geometric configuration detection technique, brightness detection technique, contour detecting technology, yardstick measure detection technique, charactron reading technology, character recognition technology, bar code identification technology.
Picture contrast, brightness regulation: by increasing integral image pixel value or reducing the brightness of image, by increasing the pixel value of the larger pixel of gray-scale value and reducing the less pixel value of gray-scale value, improve the contrast of light and shade object.Be applicable to object: picture format is U8 form, process need be strengthened to image light and shade, make light and shade excessively obvious, bright place is brighter, dark place is darker, the process of filtering intermediate value, is mainly following algorithm and provides process: technique of image edge detection, geometric configuration detection technique, brightness detection technique, contour detecting technology, charactron reading technology, character recognition technology, bar code identification technology.
Filtering: the noise produced in shooting by gaussian filtering, mean filter and median filtering algorithm removal image.Be applicable to object: picture format is U8 form, has obvious dim spot in image taking, the image of bright spot.
Binaryzation: RGB image is converted to gray level image, by comparing to pixel each in image and setting value, is 255 for the value pixel transitions being greater than setting, otherwise is 0 converted image.
Be applicable to object: picture format is U8 form, is mainly following algorithm and provides process: brightness detection technique, contour detecting technology.
Image algorithm converts: to image carry out pixel add, subtract, with or logical relation computing change.Be applicable to object: obvious not for some feature of components and parts, but carrying out with other pixel values adding, subtract, with or the image that can make moderate progress after processing.
Particle analysis: the hollow sectors that in the fine particle of object edge or blank map picture, region to be analyzed is tiny is removed to the image converting gray scale to.Be applicable to object: picture format is U8 form, comparatively sharp-pointed to image border, image inside has the image of minuscule hole to process.
Said method there is no ordinal relation, can combine as the case may be, such as, carry out the adjustment of brightness of image, contrast after gradation of image conversion, also can be the process carrying out image binaryzation after gradation of image transforms, carry out combined treatment according to specific needs.
Described image algorithm comprises:
Wherein the image algorithm of rgb format image comprises:
Color of image extraction and analysis: the light being reflected different colours by light-illuminating on object, object color is obtained by the GTG that the analysis of light extracted to often kind of light or the Cumulative sum that carries out GTG, differentiate whether the color of object meets the requirements by the color of getting, whether judgment object is because the wrong part of color or few part.
Whether color of image is in conjunction with local shape analysis: consistent with the object needed by extracting the profile combination judgment object CF excessively formed between grey decision-making and GTG reflected from object, determines wrong part, few part or reverse.
The image algorithm of U8 format-pattern comprises:
Image Edge-Detection analyze: RGB image being converted to gray level image, determining the number of excessive point, line, surface by arranging value excessive between GTG, utilize number come judgment object whether reverse or lack part;
Whether geometric configuration detects is analyzed: change RGB image into gray level image, be combined to form obvious contour of object carry out rendering image shape by the transitional region between GTG, analyzes consistent with object under test profile, and whether judgment object lacks part or reverse;
Brightness detects to be analyzed: RGB image is converted to gray level image, asks for maximal value, minimum value and accumulated value to region to be analyzed grey decision-making in image, judge whether the luminous object in analyzed region exists;
Whether less contour detecting is analyzed: RGB image being converted to gray level image, filtering out unwanted pattern by arranging value excessive between GTG, retains the pattern needed, by pattern correspondance's judgment object part or wrong part;
Yardstick is measured to detect and is analyzed: go out by image and actual photographed areal calculation the actual physical size that single pixel occupies, area, the distance of pattern is calculated by the pixel value extracting region to be analyzed in image, utilize the distance of trying to achieve or area whether normal to the distance or area that judge provincial characteristics to be analyzed, whether judgment object is reverse, wrong part or few part;
Charactron reading Analysis: by zooming in or out the feature of seven segment code in the charactron of image region to be analyzed, distinguish different display pattern, according to the content definition of different pattern being differentiated to pattern displaying, judge charactron luminescence display whether normal or charactron whether less part by the content of definition;
Text region is analyzed: extract zone map to be analyzed in image, analyzing pattern names hereof with word or indications mode, name is carried out to similar pattern and extracts word or indications, by carrying out judgment object whether less part or wrong part to the word read or indications.
Bar code recognition is analyzed: extract bar pattern, extracts bar code or Quick Response Code content according to bar code or Quick Response Code array mode.
Above-mentioned image algorithm performs successively, by extracting dissimilar feature, completes few part of product components and parts, wrong part and reverse investigation and record.
The above embodiment is only be described the preferred embodiment of the present invention; not scope of the present invention is limited; under not departing from the present invention and designing the prerequisite of spirit; the various distortion that those of ordinary skill in the art make technical scheme of the present invention and improvement, all should fall in protection domain that claims of the present invention determine.
Claims (5)
1. components and parts few part, wrong part and a reverse optical detecting method in part mate, is characterized in that, comprise the steps:
1) the rgb format image of components and parts in collected by camera plug-in unit is used;
2) Image semantic classification: to image when imaging because of extraneous factor interference cause imaging cannot show image full detail or to image in do not pay close attention to part and carry out filtering, then rgb format image is converted to U8 form;
3) feature-extraction analysis: carry out feature extraction respectively to the rgb format image of the concern part stayed after Image semantic classification and U8 format-pattern, adopts multiple image algorithm to judge few part of product components and parts, wrong part and oppositely successively, and output detections result;
4) with report form represent test every day defective products seniority among brothers and sisters, convenient storage is carried out to test data and follows the trail of inquiry in the future.
2. optical detecting method according to claim 1, is characterized in that, the method for described Image semantic classification comprises following method:
Gradation conversion: gray level image U8 form is converted into RGB image, the color that filtering is abundant, only stays black in the white and black image to white middle transition pixel value;
Picture contrast, brightness regulation: by increasing integral image pixel value or reducing the brightness of image, by increasing the pixel value of the larger pixel of gray-scale value and reducing the less pixel value of gray-scale value, improve the contrast of light and shade object;
Filtering: the noise produced in shooting by gaussian filtering, mean filter and median filtering algorithm removal image;
Binaryzation: RGB image is converted to gray level image, by comparing to pixel each in image and setting value, is 1 for the value pixel transitions being greater than setting, otherwise is 0 converted image;
Image algorithm converts: to image carry out pixel add, subtract, with or logical relation computing change;
Particle analysis: the hollow sectors that in the fine particle of object edge or blank map picture, region to be analyzed is tiny is removed to the image converting gray scale to.
3. optical detecting method according to claim 1, is characterized in that, in step 3) feature-extraction analysis, the image algorithm of described rgb format image comprises:
Color of image extraction and analysis: the light being reflected different colours by light-illuminating on object, object color is obtained by the GTG that the analysis of light extracted to often kind of light or the Cumulative sum that carries out GTG, differentiate whether the color of object meets the requirements by the color of getting, whether judgment object is because the wrong part of color or few part;
Whether color of image is in conjunction with local shape analysis: consistent with the object needed by extracting the profile combination judgment object CF excessively formed between grey decision-making and GTG reflected from object, determines wrong part, few part or reverse.
4. optical detecting method according to claim 1, is characterized in that, in step 3) feature-extraction analysis, the image algorithm of described U8 format-pattern comprises:
Image Edge-Detection analyze: RGB image being converted to gray level image, determining the number of excessive point, line, surface by arranging value excessive between GTG, utilize number come judgment object whether reverse or lack part;
Whether geometric configuration detects is analyzed: change RGB image into gray level image, be combined to form obvious contour of object carry out rendering image shape by the transitional region between GTG, analyzes consistent with object under test profile, and whether judgment object lacks part or reverse;
Brightness detects to be analyzed: RGB image is converted to gray level image, asks for maximal value, minimum value and accumulated value to region to be analyzed grey decision-making in image, judge whether the luminous object in analyzed region exists;
Whether less contour detecting is analyzed: RGB image being converted to gray level image, filtering out unwanted pattern by arranging value excessive between GTG, retains the pattern needed, by pattern correspondance's judgment object part or wrong part;
Yardstick is measured to detect and is analyzed: go out by image and actual photographed areal calculation the actual physical size that single pixel occupies, area, the distance of pattern is calculated by the pixel value extracting region to be analyzed in image, utilize the distance of trying to achieve or area whether normal to the distance or area that judge provincial characteristics to be analyzed, whether judgment object is reverse, wrong part or few part;
Charactron reading Analysis: by zooming in or out the feature of seven segment code in the charactron of image region to be analyzed, distinguish different display pattern, according to the content definition of different pattern being differentiated to pattern displaying, judge charactron luminescence display whether normal or charactron whether less part by the content of definition;
Text region is analyzed: extract zone map to be analyzed in image, analyzing pattern names hereof with word or indications mode, name is carried out to similar pattern and extracts word or indications, by carrying out judgment object whether less part or wrong part to the word read or indications.
5. optical detecting method according to claim 1, it is characterized in that, when object is for motion or meeting consecutive variations state, described optical detecting method also comprises image characteristics extraction Avi videoing skill: by arranging Image Pretreatment Algorithm in advance, carries out the reservation of processed group synthetic video hereof to every two field picture of camera shooting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510698911.0A CN105389805A (en) | 2015-10-25 | 2015-10-25 | Optical detection method of component shortage, wrong component and reverse component in plug-in operation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510698911.0A CN105389805A (en) | 2015-10-25 | 2015-10-25 | Optical detection method of component shortage, wrong component and reverse component in plug-in operation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105389805A true CN105389805A (en) | 2016-03-09 |
Family
ID=55422054
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510698911.0A Pending CN105389805A (en) | 2015-10-25 | 2015-10-25 | Optical detection method of component shortage, wrong component and reverse component in plug-in operation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105389805A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111731615A (en) * | 2020-08-21 | 2020-10-02 | 佛山隆深机器人有限公司 | Water heater packaging method and system |
| CN113759200A (en) * | 2021-09-29 | 2021-12-07 | 中国电子科技集团公司第三十八研究所 | Digital plug-in general automatic test system and method based on image processing |
| US11714402B2 (en) | 2021-09-29 | 2023-08-01 | 38Th Research Institute, China Electronics Technology Group Corporation | Universal automatic test system for digital plugboard based on imagine processing |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101915769A (en) * | 2010-06-29 | 2010-12-15 | 华南理工大学 | Automatic optical inspection method for printed circuit board comprising resistance element |
| CN102928441A (en) * | 2012-10-25 | 2013-02-13 | 渭南师范学院 | Scanning type automatic optical detecting system |
| CN102937595A (en) * | 2012-11-13 | 2013-02-20 | 浙江省电力公司电力科学研究院 | Method, device and system for detecting printed circuit board (PCB) |
| US20130163858A1 (en) * | 2011-12-21 | 2013-06-27 | Electronics And Telecommunications Research Institute | Component recognizing apparatus and component recognizing method |
| CN103279956A (en) * | 2013-05-30 | 2013-09-04 | 华南理工大学 | Method for detecting positional accuracy of components of chip mounter |
| CN103745475A (en) * | 2014-01-22 | 2014-04-23 | 哈尔滨工业大学 | Detection and positioning method used for spherical pin element |
-
2015
- 2015-10-25 CN CN201510698911.0A patent/CN105389805A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101915769A (en) * | 2010-06-29 | 2010-12-15 | 华南理工大学 | Automatic optical inspection method for printed circuit board comprising resistance element |
| US20130163858A1 (en) * | 2011-12-21 | 2013-06-27 | Electronics And Telecommunications Research Institute | Component recognizing apparatus and component recognizing method |
| CN102928441A (en) * | 2012-10-25 | 2013-02-13 | 渭南师范学院 | Scanning type automatic optical detecting system |
| CN102937595A (en) * | 2012-11-13 | 2013-02-20 | 浙江省电力公司电力科学研究院 | Method, device and system for detecting printed circuit board (PCB) |
| CN103279956A (en) * | 2013-05-30 | 2013-09-04 | 华南理工大学 | Method for detecting positional accuracy of components of chip mounter |
| CN103745475A (en) * | 2014-01-22 | 2014-04-23 | 哈尔滨工业大学 | Detection and positioning method used for spherical pin element |
Non-Patent Citations (3)
| Title |
|---|
| 吴晖辉 等: "阻容类表面贴装元件的炉前检测算法", 《测试技术学报》 * |
| 王辉: "图像处理技术在自动视觉检查系统（AOI）中的应用研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 王雁玲: "自动光学检测系统中图像处理技术研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111731615A (en) * | 2020-08-21 | 2020-10-02 | 佛山隆深机器人有限公司 | Water heater packaging method and system |
| CN113759200A (en) * | 2021-09-29 | 2021-12-07 | 中国电子科技集团公司第三十八研究所 | Digital plug-in general automatic test system and method based on image processing |
| US11714402B2 (en) | 2021-09-29 | 2023-08-01 | 38Th Research Institute, China Electronics Technology Group Corporation | Universal automatic test system for digital plugboard based on imagine processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11849998B2 (en) | Method for pupil detection for cognitive monitoring, analysis, and biofeedback-based treatment and training | |
| KR101554403B1 (en) | Image processing device, image processing method, and recording medium for control program | |
| EP2959454B1 (en) | Method, system and software module for foreground extraction | |
| CN107705288B (en) | Infrared video detection method for dangerous gas leakage under strong interference of pseudo-target motion | |
| CN103294994B (en) | For the device of face is recognized based on environmental suitability | |
| US7912251B2 (en) | Eyelid detection apparatus and program therefor | |
| CN105632015B (en) | A kind of bank money fingerprint characteristic anti false authentication method based on cloud platform | |
| CN104794491A (en) | Fuzzy clustering steel plate surface defect detection method based on pre classification | |
| CN104749184A (en) | Automatic optical detection method and automatic optical detection system | |
| CN104133548A (en) | Method and device for determining viewpoint area and controlling screen luminance | |
| CN111289538A (en) | PCB element detection system and detection method based on machine vision | |
| CN101339603A (en) | Method for selecting qualified iris image from video frequency stream | |
| CN111524101A (en) | Electronic screen defect detection method based on machine vision technology | |
| CN107909063B (en) | Biological identification video playback attack detection method based on gray level change | |
| WenJuan et al. | A real-time lip localization and tacking for lip reading | |
| CN105389805A (en) | Optical detection method of component shortage, wrong component and reverse component in plug-in operation | |
| CN106228541A (en) | Screen positioning method and device in visual inspection | |
| Sagheer et al. | An effective face detection algorithm based on skin color information | |
| CN103049748A (en) | Behavior-monitoring method and behavior-monitoring system | |
| CN117808746A (en) | Fruit quality grading method based on image processing | |
| CN108154116A (en) | A kind of image-recognizing method and system | |
| Kumar et al. | Text detection using multilayer separation in real scene images | |
| CN111160080A (en) | Intelligent automobile appearance consistency detection system based on image recognition | |
| JP2000023141A (en) | Moving object monitoring device | |
| CN111930987A (en) | Intelligent metropolitan area positioning method and system based on machine vision color recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160309 |