CN104239864A - Freight car number identification system based on image processing - Google Patents
Freight car number identification system based on image processing Download PDFInfo
- Publication number
- CN104239864A CN104239864A CN201410467777.9A CN201410467777A CN104239864A CN 104239864 A CN104239864 A CN 104239864A CN 201410467777 A CN201410467777 A CN 201410467777A CN 104239864 A CN104239864 A CN 104239864A
- Authority
- CN
- China
- Prior art keywords
- image
- car number
- character
- license number
- recognition
- 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
Landscapes
- Character Discrimination (AREA)
Abstract
The invention discloses a freight car number identification system based on image processing. The system conducts identification through the following steps that 1, a car number is positioned, wherein collected video images or static pictures are subjected to filtering, boundary enhancement and other processing, and the position of the car number is found automatically in the collected carriage image; 2, the images are preprocessed, wherein the car number image is subjected to graying, equalization, edge processing, binarization and other processing to make preparation for work in the next step; 3, characters are divided, wherein a car number area is divided into individual character areas to make preparation for character identification and solve the problems of ambiguity, character breakage and the like; 4, the characters are identified, and the process is a pattern recognition process.
Description
Technical field
The present invention relates to a kind of image procossing and mode identification method, particularly a kind of lorry car number identification system based on image procossing.
Background technology
Abroad for the research comparatively morning of railway freight-car license number, the end of the sixties, the beginning of the seventies, foreign railway system just starts to be studied car number automatic identification.Owing to affecting by technical development at that time, the technology such as usual employing color bar code, magnetic induction, surface acoustic wave realize the automatic identification of lorry, but in use, all because of the embody rule bad environments at scene, all the time fail resolution system accuracy of identification not high and interference free performance difference and wait technical barrier, so there is no be used widely.Enter the eighties, along with the development of computer technology and microwave technology, external many companies are all being devoted to adopt microwave reflection modulation technique to the research realizing locomotive, lorry identifies automatically, because technique has higher interference free performance and higher accuracy of identification, substantially solve the technical barrier affecting recognition correct rate existed in aforementioned identification technology, become the technology of current main flow.Current domestic lorry car number automatic identification also can adopt the technology based on image procossing, mainly comprises software mode and software and hardware combining mode two kinds of technical schemes.
As can be seen from the performance index of more current Train number recognition products, discrimination and the recognition speed of its system have much room for improvement, and in license number location and identification, also do not have a kind of general method, location and recognizer remain a current research direction that need solve at a high speed, accurately in research.For location and the identification of lorry license number, there is a lot of difficult points.
Summary of the invention
A kind of lorry car number identification system based on image procossing of the present invention, it is exactly mainly the demand for current international and domestic TMIS develop rapidly, construct the lorry car number identification system framework based on image procossing, a kind of digital image processing techniques that utilize are proposed to research and develop the method for railway freight-car ATIS, provide a method effectively automatically can located, split, identify railway freight-car license number, actual railway administration system can be applied to, and further raising can be obtained on discrimination and recognition speed.
The object of the invention is to be realized by several like this step like this:
1, license number location: the process such as filtering, border enhancing is carried out to the video image collected or static images, be automatically found the position of license number in the compartment image collected;
2, Image semantic classification: license number image is carried out to the process such as gray processing, equalization, edge treated, binaryzation, for next step work is ready;
3, Character segmentation: license number region segmentation is become single character zone, for character recognition is prepared, overcomes the problems such as fuzzy, character fracture;
4, character recognition: this process is exactly pattern-recognition.
The present invention can also comprise:
A () adopts the method edge image expanded to process, the defined formula of expansion:
(1);
Wherein, A is called input picture, and B is called structural element.Dilation operation can fill and lead up the small gap of aperture in image and fracture.
B after dilation operation, also can there are some little noise and burrs in image in (), so, morphologic corrosion can be utilized to carry out noise reduction process to image.The defined formula of corrosion is as follows:
(2)。
Accompanying drawing explanation
Fig. 1 is the procedural block diagram of Train number recognition.
Embodiment
Below in conjunction with accompanying drawing citing, the present invention is described in more detail:
Embodiment 1
Composition graphs 1, Fig. 1 illustrates the simple framework of car number identification system, and main utilization is programmed based on the digital image processing techniques of visual c++, processes, completed gray processing, edge extracting, binaryzation, slant correction scheduling algorithm involved in pre-service.And obtain good result.
The present invention adopts method of weighted mean to carry out gray proces, because R, G, B are on the impact of image color and be not quite similar, and the mankind are least responsive to B, slightly strong to R, the most responsive to G, other experiment conclusion according to some again, we give different weights Wr, Wg, Wb can to R, G, B, then use equation below to it: also i.e. R=G=B=(WrR+WgG+WbB)/3.By years of researches and practice, it is found that the most rational weights are Wr=0.30, namely Wg=0.59, Wb=0.11. work as Vgray=0.30R+0.59G+0.11B; During R=G=B=Vgray, the gray level image obtained is the most reasonable.
the present invention adopts Roberts operator to carry out edge treated to license number image, and Roberts edge detection operator very successfully can extract the edge of object, has formula below to provide
(3);
Present invention employs Global thresholding and binaryzation is carried out to license number image.Choosing of threshold value is the key of result quality, adopts solution by iterative method to publish picture the optimal threshold of picture.Its concrete grammar is: (a) obtains the grey level histogram of license number image; (b) according to the trough value of grey level histogram using its initial value as iteration.Suppose that image gray levels is L, so:
(4);
Wherein, the number of pixels of hl to be gray-scale value be l.Iteration is performed until Ti+1=Ti and terminates, and the Ti at the end of getting is final segmentation threshold T.Due to the singularity (background environment is more complicated, and dust covers serious, and license number has stained etc.) of lorry license plate image, in order to obtain good treatment effect, K is usually wanted to be in the state of dynamic change.First we first can get an initial value K0, and ensuing choosing can according to following experimental knowledge, that is: in license number image, background pixel point sum is substantially fixing with the ratio of character pixels point sum.
(5);
Although the license plate image of goods train has the feature of itself, affected by environment also comparatively large, S value has certain general variation range, if at S
0fluctuate, so when carrying out interative computation at every turn, utilize the such characteristic of S dynamically can adjust K, the explanation as below:
Embodiment 2
Based on a lorry car number identification system for image procossing, this recognition system adopts following steps to identify:
1, license number location: the process such as filtering, border enhancing is carried out to the video image collected or static images, be automatically found the position of license number in the compartment image collected;
2, Image semantic classification: license number image is carried out to the process such as gray processing, equalization, edge treated, binaryzation, for next step work is ready;
3, Character segmentation: license number region segmentation is become single character zone, for character recognition is prepared, overcomes the problems such as fuzzy, character fracture;
4, character recognition: this process is exactly pattern-recognition;
The present invention can also comprise:
A () adopts the method edge image expanded to process, the defined formula of expansion:
(1),
Wherein, A is called input picture, and B is called structural element, and dilation operation can fill and lead up the small gap of aperture in image and fracture.
Described after dilation operation, also can there are some little noise and burrs in image, so, morphologic corrosion can be utilized to carry out noise reduction process to image, and the defined formula of corrosion is as follows:
(2)。
Claims (2)
1. based on a lorry car number identification system for image procossing, it is characterized in that: this recognition system adopts following steps to identify:
1, license number location: the process such as filtering, border enhancing is carried out to the video image collected or static images, be automatically found the position of license number in the compartment image collected;
2, Image semantic classification: license number image is carried out to the process such as gray processing, equalization, edge treated, binaryzation, for next step work is ready;
3, Character segmentation: license number region segmentation is become single character zone, for character recognition is prepared, overcomes the problems such as fuzzy, character fracture;
4, character recognition: this process is exactly pattern-recognition;
The present invention can also comprise:
The method edge image expanded is adopted to process, the defined formula of expansion:
,
Wherein, A is called input picture, and B is called structural element, and dilation operation can fill and lead up the small gap of aperture in image and fracture.
2. a kind of lorry car number identification system based on image procossing according to claim 1, it is characterized in that: described after dilation operation, also can there are some little noise and burrs in image, morphologic corrosion can be utilized to carry out noise reduction process to image, and the defined formula of corrosion is as follows:
。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410467777.9A CN104239864A (en) | 2014-09-16 | 2014-09-16 | Freight car number identification system based on image processing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410467777.9A CN104239864A (en) | 2014-09-16 | 2014-09-16 | Freight car number identification system based on image processing |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104239864A true CN104239864A (en) | 2014-12-24 |
Family
ID=52227890
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410467777.9A Pending CN104239864A (en) | 2014-09-16 | 2014-09-16 | Freight car number identification system based on image processing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104239864A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104408438A (en) * | 2014-12-10 | 2015-03-11 | 黑龙江真美广播通讯器材有限公司 | Truck number identification system |
CN106056730A (en) * | 2016-08-09 | 2016-10-26 | 南京信息工程大学 | Community garage intelligent access control system based on license plate detection and application method thereof |
US10706330B2 (en) | 2015-10-01 | 2020-07-07 | Intellivision Technologies Corp | Methods and systems for accurately recognizing vehicle license plates |
US11587327B2 (en) | 2015-10-01 | 2023-02-21 | Intellivision Technologies Corp | Methods and systems for accurately recognizing vehicle license plates |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6281928B1 (en) * | 1998-05-13 | 2001-08-28 | Chuo Hatsujo Kabushiki Kaisha | Positional detector device for a vehicular license plate |
CN102708356A (en) * | 2012-03-09 | 2012-10-03 | 沈阳工业大学 | Automatic license plate positioning and recognition method based on complex background |
CN103065138A (en) * | 2012-12-06 | 2013-04-24 | 新疆公众信息产业股份有限公司 | Recognition method of license plate number of motor vehicle |
CN104680199A (en) * | 2013-11-27 | 2015-06-03 | 哈尔滨功成科技创业投资有限公司 | System for identifying freight car number based on image processing |
-
2014
- 2014-09-16 CN CN201410467777.9A patent/CN104239864A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6281928B1 (en) * | 1998-05-13 | 2001-08-28 | Chuo Hatsujo Kabushiki Kaisha | Positional detector device for a vehicular license plate |
CN102708356A (en) * | 2012-03-09 | 2012-10-03 | 沈阳工业大学 | Automatic license plate positioning and recognition method based on complex background |
CN103065138A (en) * | 2012-12-06 | 2013-04-24 | 新疆公众信息产业股份有限公司 | Recognition method of license plate number of motor vehicle |
CN104680199A (en) * | 2013-11-27 | 2015-06-03 | 哈尔滨功成科技创业投资有限公司 | System for identifying freight car number based on image processing |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104408438A (en) * | 2014-12-10 | 2015-03-11 | 黑龙江真美广播通讯器材有限公司 | Truck number identification system |
US10706330B2 (en) | 2015-10-01 | 2020-07-07 | Intellivision Technologies Corp | Methods and systems for accurately recognizing vehicle license plates |
US11587327B2 (en) | 2015-10-01 | 2023-02-21 | Intellivision Technologies Corp | Methods and systems for accurately recognizing vehicle license plates |
CN106056730A (en) * | 2016-08-09 | 2016-10-26 | 南京信息工程大学 | Community garage intelligent access control system based on license plate detection and application method thereof |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2022027931A1 (en) | Video image-based foreground detection method for vehicle in motion | |
US20210049769A1 (en) | Vibe-based three-dimensional sonar point cloud image segmentation method | |
CN105374015A (en) | Binary method for low-quality document image based on local contract and estimation of stroke width | |
CN103413276A (en) | Depth enhancing method based on texture distribution characteristics | |
CN104239864A (en) | Freight car number identification system based on image processing | |
CN107085836B (en) | Universal ghost eliminating method in moving object detection | |
CN112528868B (en) | Illegal line pressing judgment method based on improved Canny edge detection algorithm | |
CN105447489B (en) | A kind of character of picture OCR identifying system and background adhesion noise cancellation method | |
CN110674812B (en) | Civil license plate positioning and character segmentation method facing complex background | |
CN112200742A (en) | Filtering and denoising method applied to edge detection | |
Kortli et al. | Efficient implementation of a real-time lane departure warning system | |
CN111369570A (en) | Multi-target detection tracking method for video image | |
Ashoori-Lalimi et al. | An efficient method for vehicle license plate detection in complex scenes | |
CN102184553B (en) | Moving shadow detecting method and device | |
CN104657969A (en) | Method and system for removing image background | |
CN109636822A (en) | A kind of improvement Canny adaptive edge extracting method based on new building subordinating degree function | |
Dev | A novel approach for car license plate detection based on vertical edges | |
CN104102911A (en) | Image processing for AOI (automated optical inspection)-based bullet appearance defect detection system | |
CN104680199A (en) | System for identifying freight car number based on image processing | |
CN108009480A (en) | A kind of image human body behavioral value method of feature based identification | |
CN106951831B (en) | Pedestrian detection tracking method based on depth camera | |
Wenzhong et al. | A watershed based segmentation method for overlapping chromosome images | |
CN104408438A (en) | Truck number identification system | |
CN102034227A (en) | Method for de-noising image | |
CN108961290B (en) | Adaptive image edge detection method based on Ratio operator of Otsu |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20141224 |
|
WD01 | Invention patent application deemed withdrawn after publication |