CN102682267B - Method for quickly positioning and extracting one-dimensional and two-dimensional code graphics aiming at digital images - Google Patents
Method for quickly positioning and extracting one-dimensional and two-dimensional code graphics aiming at digital images Download PDFInfo
- Publication number
- CN102682267B CN102682267B CN201210157354.8A CN201210157354A CN102682267B CN 102682267 B CN102682267 B CN 102682267B CN 201210157354 A CN201210157354 A CN 201210157354A CN 102682267 B CN102682267 B CN 102682267B
- Authority
- CN
- China
- Prior art keywords
- region
- quick response
- dimensional
- response code
- code
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a method for quickly positioning and extracting one-dimensional and two-dimensional code graphics aiming at digital images. The method comprises the following steps of: analyzing envelopes of a high-frequency component for observing an image signal, extracting a region with an obvious upper and lower envelope difference under each scale as a probable bar code region, further analyzing the probable bar code region, extracting geometric parameters from partial information and performing adaptive binarization, supplying a result to a subsequent scanning module, and sampling and decoding to output information stored by a bar code according to a geometric relation of a black and white module. Equipment required in the method is relatively simple and can be suitable for complicated backgrounds and illumination conditions; and the observed one-dimensional and two-dimensional code graphics are quickly and intelligently processed on the basis of acquisition of video signals.
Description
Technical field
The present invention relates to Computer Image Processing method, particularly adopt the method for computer image processing technology for the automatic detection of a peacekeeping two-dimensional bar code and rapid extraction.It belongs to the preprocess method before bar-code identification, can in the image of taking, locate fast bar code region, and then be beneficial to Fast Segmentation and identification, to adopt corresponding bar-code identification algorithm, under statistical significance, accelerates system response time.
Background technology
The fast development of society in recent years and the progressive peacekeeping two-dimensional bar code that impelled are the widespread use in commercial production and daily life of gordian technique as Internet of Things field, are the information medias between the smart machines such as people, goods and computing machine.Compare with other recognition means, barcode technology has that cost is low, easy to use, high reliability.Traditional barcode recognition equipment adopts laser scanning methods.But along with the cost of visual light imaging sensor is more and more lower, and the unrivaled abundant information amount of image information, current development trend is the digital image sensor adopting based on visible images technology, the low cost solution gathering as a peacekeeping two-dimensional bar code.Another trend is that various two-dimensional bar codes are paid attention to gradually in recent years, and start to be applied to the various intelligent terminals with video capability, for example special logistics management mobile terminal or smart mobile phone.In the use procedure of barcode technology, first establish and need the information limit of transmitting to work out corresponding bar code; Then two-dimensional bar code can be printed on to the specific region that is applicable to reading, in the external packing of article; In Information Communication process, Quick Response Code is as carrier and the transmission interface of information, for example, in the outbound of article, transport, distribution, storage, transfer etc. link, can read the code that article are corresponding by scanning and obtain fast relevant information.In the use, the data volume of bar code storage is limited, therefore relies on online database and connects to retrieve detailed code information; And the information storage capability of Quick Response Code is stronger, can select according to concrete applied environment the use of online or off-line.Need emphasize, developing rapidly of barcode technology causes barcode types numerous in recent years, and its standard is different.Use single equipment to read efficiently multiple bar code, the two-dimensional bar code of especially various complexity, has certain technical difficulty simultaneously.
In conventionally processing, according to the national standard of each bar code, the identification of certain barcode types realizes by scanning and identify its positioning pattern.But, while processing multiple barcode types at the same time, scan respectively the mode of multiple Bar code positioning figure by the multiple scanning relating to for full image, cause efficiency lower, recognition speed is slack-off.This be because, be subject to the uncertainty impact of shooting condition, bar code graphics may appear at the arbitrary region in digital picture with various yardsticks, therefore must carry out exhaustive search to the each details in image, could detect the bar code graphics that may exist.Although the quantity of information of digital picture is abundant, also causes treatment effeciency not ignore.On objective demand, need to there is more efficient and unified bar code graphics Check processing pattern.How to process multiple bar codes simultaneously and detect, do not have relevant criterion technology to instruct.In fact, differentiate that to have or not certain bar code graphics be classification problem more than in observed image arbitrarily, this is a generally acknowledged difficult problem at area of pattern recognition.In addition, because bar code scanning is subject to the impact of direction and pattern distortion larger; Be subject to the restriction of ambient lighting power and degree of uniformity due to the effect of image acquisition simultaneously.But the circumstance complication of actual use procedure is changeable, often occur that two-dimensional bar is blocked and stained situation.It is more difficult that these actual conditions all cause barcode types to detect.Therefore, development detects the fast method of Quick Response Code figure under actual complex environment from observed image, has positive effect.
At home up to the present, not yet retrieve similar techniques or Patents.
Summary of the invention
The object of this invention is to provide a kind of bar code and two-dimensional bar code location and extracting method fast, to process as the early stage of barcode types identification, when solution is identified multiple barcode types simultaneously, exhaustive detects and causes the problem that recognition speed is low.
The object of the present invention is achieved like this: an a kind of peacekeeping Quick Response Code figure for digital picture is location and extracting method fast, and processing procedure is carried out according to following steps:
1), read coloured image and be converted into gray level image by computer drives digital camera;
2), gray level image is carried out to smoothing denoising pre-service;
3), gray level image is carried out to the yardstick pyramid decomposition of 4 levels;
4), the picture signal of different scale is calculated to upper lower envelope;
5), calculate the difference of lower envelope in the picture signal of different scale, the envelope difference under different scale is carried out to information fusion, obtain gray scale rank entirety disparity map;
6), calculate the marking area in the entirety disparity map of gray scale rank by maximum difference method between class, and carry out binary conversion treatment by calculating gained threshold value;
7), adopt extraction algorithm to detect to the connected component region after binaryzation, add up area the sequence of protruding connected component, and then obtain the protruding connected component region of maximum area; If the area of this connected component is greater than setting threshold, can thinks that this region is possible one dimension code and Quick Response Code region, and proceed to step 8); Otherwise think and one dimension code and Quick Response Code region do not detected, proceed to step 11);
8), calculate the geometric parameter in the protruding connected component of maximum area region, comprise boundary rectangle and fitted ellipse;
9), adopt local segmentation method to carry out binarization segmentation to the gray level image of original input to circumscribed rectangular region, and other regions outside circumscribed rectangular region are set for blank, obtain local segmentation image, and be applied to follow-up bar code information identification;
10), the binaryzation figure of acquisition being transferred to computing machine identifies; Successfully open user's prompt facility if identify, notify user and decoding information is outputed to buffer zone;
11), wait for user instruction, if receive continue instruction return to step 1); Otherwise enter step 12);
12), exit.
The present invention is made up of three main parts: have the video image acquisition part of prompting and light filling function, process the core algorithm part of a peacekeeping two-dimensional bar code extraction of video image, be responsible for the interface part with user interactions.Video acquisition part, according to user instruction, is opened camera and is gathered color video frequency image according to fixing frame per second; One peacekeeping Quick Response Code detects identification division the video image of camera acquisition is carried out to scan process, extracts and export a peacekeeping two-dimensional barcode information wherein; Man-machine interaction part receives after user's work order, first opens light compensating apparatus prompting subscriber equipment and is transferred to and being started working by dormancy, and by laser pilot light for localization prompting scanning input scope, after scanning successfully, pointed out with hummer.Three parts have comprised respectively bottom hardware and driving, core processing algorithm software and upper strata interactive software.
The present invention analyzes by the envelope of the radio-frequency component to observation image signal, being extracted in the significant region of upper and lower envelope difference under each yardstick further analyzes as possible bar code region, to local information extraction geometric parameter and self-adaption binaryzation, and then result is offered to follow-up scan module, thereby according to the geometric relationship sampling of black and white module decoding output bar code canned data.
The present invention can be integrated in special bar code reading facility, also can utilize the video acquisition hardware device resources of existing widely used embedded intelligent equipment, on the basis of video signal collective, the one dimension code observing and Quick Response Code figure be carried out to intelligentized processing.
Than the traditional standard detection method of the Quick Response Code of bar code, beneficial effect of the present invention is:
1, the present invention can adapt to one dimension code or Quick Response Code type are unrestricted, even comprise UNKNOWN TYPE;
2, the present invention can adapt to the angle of bar code in observation or the wide variation of yardstick;
3, the present invention can one-dimensional or the region of interest ROI at two-dimensional bar code place, and is supported in area-of-interest inside and carries out partial operation, to reduce traversal scope, accelerates travelling speed;
4, the present invention can carry out rude classification to barcode types, to accelerate follow-up identifying;
5, image binaryzation is integrated into detection by the present invention, can effectively strengthen efficiency;
6, can not there is not complicated plant equipment in the present invention, can make failure rate lower;
7, the present invention can adapt to complicated background and illumination condition, and the requirement of light filling is reduced, and can more be adapted to more common intelligent movable equipment.
The present invention can adapt to the light conditions under truth, for example illumination deficiency, uneven illumination are even, background is mixed and disorderly etc., and automatically detect the Quick Response Code figure in video flowing, extract its geometric parameter, local conversion is binary image efficiently again, wherein the dark striped of Quick Response Code is treated to black, and light striped and background are treated to white.Binary image can be for Quick Response Code scan module, to detect positioning pattern wherein.Whole processing procedure has been used some accelerating algorithm, makes implementation procedure higher than the speed of classic method, is not simultaneously subject to the impact of version, printed dimensions and color, and has the feature of higher antinoise and stained ability.
Brief description of the drawings
Fig. 1 is hardware connection mode key diagram of the present invention.
Fig. 2-1 and Fig. 2-2 are respectively the process flow diagrams that the present invention processes overview flow chart and the processing of multi-resolution adaptive threshold calculations.
Fig. 3-1st, video camera scheme of installation of the present invention.
Fig. 3-2nd, the another kind of scheme of installation of video camera of the present invention.
Embodiment
Basic thought of the present invention is as follows: a kind of bar code and two-dimensional bar code method for quick based on digital image processing techniques, referring to Fig. 1, the present invention is mainly made up of digital figure sensor, range sensor light filling equipment, location indicating equipment and embedded computer or industrial control computer.Computing machine directly reads Digital Image Data by driving digital camera, then transfers data to the Intelligent Measurement module of a peacekeeping Quick Response Code: carry out following detecting step:
1), the image of read sensor sampling;
2), coloured image is converted to gray level image;
3), image is carried out to the pre-service such as smoothing denoising;
4), the picture signal of different scale is calculated to upper lower envelope;
5), calculate the difference of lower envelope in the picture signal of different scale, the envelope difference under different scale is carried out to information fusion, obtain gray scale rank entirety disparity map;
6), calculate the marking area in the entirety disparity map of gray scale rank by maximum difference method between class, and carry out binary conversion treatment by calculating gained threshold value;
7), adopt extraction algorithm to detect to the connected component region after binaryzation, add up area the sequence of protruding connected component, and then obtain the protruding connected component region of maximum area; If the area of this connected component is greater than setting threshold, can thinks that this region is possible one dimension code and Quick Response Code region, and proceed to step 8); Otherwise think and one dimension code and Quick Response Code region do not detected, proceed to step 11);
8), calculate the geometric parameter in the protruding connected component of maximum area region, comprise boundary rectangle and fitted ellipse etc.;
9), adopt local segmentation method to carry out binarization segmentation to the gray level image of original input to circumscribed rectangular region, and other regions outside circumscribed rectangular region are set for blank, obtain local segmentation image, can be applicable to follow-up bar code information identification;
10), the binaryzation figure of acquisition being transferred to scanning detection apparatus identifies; Successfully open user's prompt facility if identify, notify user and decoding information is outputed to buffer zone;
11), wait for user instruction, if receive continue instruction return to step 1); Otherwise enter step 12);
12), exit.
Before step 1) starts, light filling equipment starts light filling, location indicating equipment prompting sweep limit, and detect observation content whether in the suitable distance range of 10cm – 30cm with range sensor, if distance is applicable to, proceed to step 1) (this suitable distance range refers to that it is 10cm – 30cm that barcode scanner is suitable for service range).Light filling equipment is multiple LED light compensating lamps that annular arranges.
In above each step, central idea is to adopt the mode of texture to differentiate doubtful one dimension code and Quick Response Code region, because the region of one dimension code and Quick Response Code always exists high-frequency dark color and light color conversion, and dark and light ratio is close.If successfully judgement exists Quick Response Code, take multiresolution to extract one dimension code or Quick Response Code graphics field.Due to adaptive multiresolution processing, the texture of one dimension code or Quick Response Code is being had outstanding performance compared with under large scale conventionally, therefore needn't carry out exhaustive search to whole image, computation complexity greatly reduces, and has accelerated one dimension code and Quick Response Code Detection and Extraction process (referring to Fig. 2-1, Fig. 2-2).Therefore, judge fast in observed image and whether have the meaning of one dimension code and Quick Response Code figure Fast Segmentation to be, can improve the speed that detects identification.Whether first judge in scene and have the calculation cost of one dimension code and Quick Response Code smaller, if there is no one dimension code and Quick Response Code in observed image, can avoid follow-up useless operation, direct circulation is processed next frame image.Secondly,, if having one dimension code and Quick Response Code in scene, need one dimension code and Quick Response Code graphics field to extract.Key request is not now that processing speed is fast, but accurately obtains the high-quality image of cutting apart as far as possible, and then reduces the number of times that scanning decoding module is attempted.In other words, although high-quality cutting procedure can suitably increase operand, owing to having improved hit rate, therefore see to remain from the overall situation and improved recognition rate.Because we have also adopted the method for partial high-precision binaryzation, therefore make computation burden further to reduce.
Hardware list is as follows:
Title | Model |
Digital camera | Logitech C160 |
Camera lens | Be equivalent to the 85mm optical lens of 135 cameras |
Display screen | 7’ WVGA Touchscreen LCD display (add-on module)2 LVDS connectors |
Embedded computer | Freescale i.MX51 EVK |
Light compensating apparatus | LED annular light compensating lamp |
User's suggestion device | FM hummer |
Range sensor | KEYES MMS103H |
Location indicating equipment | Green LED lamp and optically focused prism |
Camera parameters explanation (referring to Fig. 3-1, Fig. 3-2): video camera imaging effect is affected by sensitive chip and camera lens two aspects mainly.Wherein, sensitive chip can adopt CCD chip also can use CMOS chip, need adopt large-size as far as possible, to be lowered into picture noise and to improve dynamic range.Camera lens adopts tight shot, and focal length adopts the 85mm optical lens that is equivalent to 135 cameras, is consistent so that take distance and the use habit of QR code.Camera lens, without automatic focusing, only needs the suitable field depth of design, and while making hand-held or location scanning, the image of QR is just in time within the spiritual scope of camera lens.Sensitive chip, the camera lens of video camera need be relatively fixing, are integrated together.
Hardware connection description: camera lens keeps constant with the relative position of camera sensitive chip by stationary installation.The signal of sensitive chip is exported to one dimension code and Quick Response Code detection computations machine after converting by built-in AD, and computing machine connects and controls light filling equipment and positioning equipment, and display screen and hummer are also connected to computing machine by slot simultaneously.
Fig. 3-1 illustrates, camera 2 is arranged on mobile computer 1 front portion.
Fig. 3-2 illustrate, camera 3 is arranged on the top board of worktable 4, and desk-top computer 5 is arranged on the support side of worktable.
Claims (3)
1. for peacekeeping Quick Response Code figure location and an extracting method fast for digital picture, it is characterized in that, processing procedure is carried out according to following steps:
1), read coloured image and be converted into gray level image by computer drives digital camera;
2), gray level image is carried out to smoothing denoising pre-service;
3), gray level image is carried out to the yardstick pyramid decomposition of 4 levels;
4), the picture signal of different scale is calculated to upper lower envelope;
5), calculate the difference of lower envelope in the picture signal of different scale, the envelope difference under different scale is carried out to information fusion, obtain gray scale rank entirety disparity map;
6), calculate the marking area in the entirety disparity map of gray scale rank by maximum difference method between class, and carry out binary conversion treatment by calculating gained threshold value;
7), adopt extraction algorithm to detect to the connected component region after binaryzation, add up area the sequence of protruding connected component, and then obtain the protruding connected component region of maximum area; If the area of this connected component is greater than setting threshold, can thinks that this region is possible one dimension code and Quick Response Code region, and proceed to step 8); Otherwise think and one dimension code and Quick Response Code region do not detected, proceed to step 11);
8), calculate the geometric parameter in the protruding connected component of maximum area region, comprise boundary rectangle and fitted ellipse;
9), adopt local segmentation method to carry out binarization segmentation to the gray level image of original input to circumscribed rectangular region, and other regions outside circumscribed rectangular region are set for blank, obtain local segmentation image, and be applied to follow-up bar code information identification;
10), the binaryzation figure of acquisition being transferred to computing machine identifies; Successfully open user's prompt facility if identify, notify user and decoding information is outputed to buffer zone;
11), wait for user instruction, if receive continue instruction return to step 1); Otherwise enter step 12);
12), exit.
2. an a kind of peacekeeping Quick Response Code figure for digital picture according to claim 1 location and extracting method fast, it is characterized in that, before described step 1) starts, light filling equipment starts light filling, location indicating equipment prompting sweep limit, and detect observation content whether in the suitable distance range of 10cm – 30cm with range sensor, if distance is applicable to, proceed to step 1).
3. an a kind of peacekeeping Quick Response Code figure for digital picture according to claim 2 location and extracting method fast, is characterized in that, described light filling equipment is multiple LED light compensating lamps that annular arranges.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210157354.8A CN102682267B (en) | 2012-05-18 | 2012-05-18 | Method for quickly positioning and extracting one-dimensional and two-dimensional code graphics aiming at digital images |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210157354.8A CN102682267B (en) | 2012-05-18 | 2012-05-18 | Method for quickly positioning and extracting one-dimensional and two-dimensional code graphics aiming at digital images |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102682267A CN102682267A (en) | 2012-09-19 |
CN102682267B true CN102682267B (en) | 2014-07-16 |
Family
ID=46814168
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210157354.8A Active CN102682267B (en) | 2012-05-18 | 2012-05-18 | Method for quickly positioning and extracting one-dimensional and two-dimensional code graphics aiming at digital images |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102682267B (en) |
Families Citing this family (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103177416B (en) * | 2013-04-10 | 2015-08-19 | 浙江大学 | A kind of QR code image position method based on least square method |
CN104424458B (en) * | 2013-08-23 | 2017-08-04 | 希姆通信息技术(上海)有限公司 | Image-recognizing method and device, the webserver, image recognition apparatus and system |
CN103997603A (en) * | 2014-03-25 | 2014-08-20 | 苏州吉视电子科技有限公司 | Configuring method of intelligent camera |
CN104200189B (en) * | 2014-08-27 | 2017-05-03 | 苏州佳世达电通有限公司 | Barcode scanning device and processing method thereof |
CN104732183B (en) * | 2015-03-20 | 2017-06-13 | 杭州晟元数据安全技术股份有限公司 | A kind of bar code recognition methods based on the analysis of image sampling line half-tone information |
CN105609008A (en) * | 2016-02-05 | 2016-05-25 | 中国科学院理化技术研究所 | Control system based on two-dimension code images and control method based on two-dimension code images |
CN105787409A (en) * | 2016-02-20 | 2016-07-20 | 深圳市欣视景科技股份有限公司 | Express delivery package identification code scanner |
CN105975894B (en) * | 2016-05-11 | 2018-08-21 | 华中科技大学 | A kind of one-dimension code recognizer based on auto-adaptable image edge detection and mapping model |
CN106201363B (en) * | 2016-07-26 | 2023-01-31 | 四川大学 | Memory and storage method for random real-time access of video stream pixel-level data |
CN107665324B (en) * | 2016-07-27 | 2020-08-28 | 腾讯科技(深圳)有限公司 | Image identification method and terminal |
CN106485184B (en) * | 2016-11-29 | 2018-12-14 | 北京慧眼智行科技有限公司 | A kind of barcode scanning method and apparatus |
CN109800615A (en) * | 2018-12-28 | 2019-05-24 | 新大陆数字技术股份有限公司 | The detection localization method and system of two-dimension code area |
CN109993018A (en) * | 2019-04-04 | 2019-07-09 | 哈尔滨理工大学 | It is a kind of based on the two dimensional code identifying system of Zynq heterogeneous platform and recognition methods |
CN110728163B (en) * | 2019-10-11 | 2023-05-09 | 首都医科大学宣武医院 | Two-dimensional code authenticated through opaque medium and authentication system thereof |
CN111191480B (en) * | 2019-12-30 | 2023-08-29 | 广东旭龙物联科技股份有限公司 | Method and device for searching bar code boundary by envelope curve based on first-order difference |
CN112651257B (en) * | 2020-12-23 | 2022-08-16 | 福建新大陆支付技术有限公司 | Two-dimensional code, bar code image positioning and identifying method and storage medium thereof |
CN115270837B (en) * | 2022-09-27 | 2023-01-06 | 北京紫光青藤微系统有限公司 | One-dimensional code identification device and identification method |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101093544A (en) * | 2007-06-14 | 2007-12-26 | 中兴通讯股份有限公司 | Method for correcting pattern in 2D code of perspective quick response matrix |
-
2012
- 2012-05-18 CN CN201210157354.8A patent/CN102682267B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101093544A (en) * | 2007-06-14 | 2007-12-26 | 中兴通讯股份有限公司 | Method for correcting pattern in 2D code of perspective quick response matrix |
Non-Patent Citations (6)
Title |
---|
Automatic Real-Time Barcode Localization in Complex Scenes;Chunhui Zhang et al;《2006 IEEE International Conference on Image Processing》;20060811;497-500 * |
Chunhui Zhang et al.Automatic Real-Time Barcode Localization in Complex Scenes.《2006 IEEE International Conference on Image Processing》.2006,497-500. |
基于多方法融合的文本定位算法的研究;潘道远等;《计算机应用与软件》;20100630;第27卷(第6期);236-238、255 * |
潘道远等.基于多方法融合的文本定位算法的研究.《计算机应用与软件》.2010,第27卷(第6期),236-238、255. |
移动端二维条码图像增强及应用研究;蔡文婷;《浙江工业大学》;20090615;1-70 * |
蔡文婷.移动端二维条码图像增强及应用研究.《浙江工业大学》.2009,1-70. |
Also Published As
Publication number | Publication date |
---|---|
CN102682267A (en) | 2012-09-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102682267B (en) | Method for quickly positioning and extracting one-dimensional and two-dimensional code graphics aiming at digital images | |
CN102393906B (en) | Computer vision inspection technology-based method for rapid detection and extraction of two-dimensional code | |
CN102231187B (en) | Computer vision detection technology-based method for detecting and identifying QR (Quick Response) code | |
US7613328B2 (en) | Label detection | |
US10558844B2 (en) | Lightweight 3D vision camera with intelligent segmentation engine for machine vision and auto identification | |
CN103295034B (en) | A kind of embedded Books Chaotic Frame checking system based on DSP and method | |
CN202632312U (en) | System based on image identification technology to collecting and processing bar code information at the same time | |
CN108982512A (en) | A kind of circuit board detecting system and method based on machine vision | |
CN105427320A (en) | Image segmentation and extraction method | |
CN111222382A (en) | Commodity settlement method, commodity settlement device, commodity settlement medium and electronic equipment based on images | |
CN111259908A (en) | Machine vision-based steel coil number identification method, system, equipment and storage medium | |
CN104463067B (en) | Method for extracting macro blocks of Grid Matrix two-dimensional bar code | |
CN105162520A (en) | Automatic identification method and information service system based on visible light illumination | |
CN110008771B (en) | Code scanning system and code scanning method | |
CN101930532A (en) | Mobile phone-based camera quick response matrix code recognizing and reading method | |
CN104200249A (en) | Automatic clothes matching method, device and system | |
CN111881832A (en) | Lane target detection method, device, equipment and computer readable storage medium | |
CN107014818A (en) | A kind of laser carving defective vision detecting system and method | |
CN101937505B (en) | Target detection method and equipment and used image acquisition device thereof | |
Jin et al. | A Smart Water Metering System Based on Image Recognition and Narrowband Internet of Things. | |
CN102610104A (en) | Onboard front vehicle detection method | |
CN116524195A (en) | Semantic segmentation method, semantic segmentation device, electronic equipment and storage medium | |
CN104715250B (en) | cross laser detection method and device | |
CN115810197A (en) | Multi-mode electric power form recognition method and device | |
WO2020083283A1 (en) | Payment method capable of automatically recognizing payment amount |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |