CN108009545A - A kind of automobile engine cylinder-body sequence number vision automatic recognition method - Google Patents
A kind of automobile engine cylinder-body sequence number vision automatic recognition method Download PDFInfo
- Publication number
- CN108009545A CN108009545A CN201711408666.0A CN201711408666A CN108009545A CN 108009545 A CN108009545 A CN 108009545A CN 201711408666 A CN201711408666 A CN 201711408666A CN 108009545 A CN108009545 A CN 108009545A
- Authority
- CN
- China
- Prior art keywords
- image
- engine cylinder
- automobile engine
- sequence number
- profile
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/09—Recognition of logos
Abstract
A kind of automobile engine cylinder-body sequence number vision automatic recognition method, belongs to sequence number identification technology field.Technical solution:All model images are read in, binary conversion treatment is carried out to image, profile is retrieved from bianry image;The automobile engine cylinder-body serial number image for needing to identify is read in, binary conversion treatment is carried out to image, profile is retrieved from bianry image;Circulate automobile engine cylinder-body serial number image profile information, detect in the circulating cycle, cutting numeral/alphabetical profile, it is in the same size with model image, image after cutting subtracts each other with model image respective pixel point value, the quadratic sum of the pixel point value of the gained whole picture of picture is calculated, one group of corresponding model image of the quadratic sum minimum is numeral/letter to be identified.Beneficial effect is:Automobile engine cylinder-body sequence number vision automatic recognition method of the present invention the method increase the stability and high efficiency, the various defects for overcoming artificial handwork of operation, save operating personnel's quantity.
Description
Technical field
The invention belongs to sequence number identification technology field, more particularly to a kind of automobile engine cylinder-body sequence number vision are automatic
Recognition methods.
Background technology
Automobile engine cylinder-body sequence number is engine production enterprise according to pertinent regulations, enterprise or industrial practice and hair
The attribute of motivation, is the cognizance code of certain a collection of like products establishment, to represent the manufacturing enterprise of engine, specification, performance,
The relevant informations such as feature, technique, purposes and product batches.Such as fuel type, number of cylinders, discharge capacity and brake stator power.With
Past automobile engine cylinder-body sequence number identification can only rely on manual work, and operation craftization, task difficulty are big, and manual work is easy
Visual fatigue, easy maloperation.
The content of the invention
In order to solve to rely on manual work in the prior art, operation craftization, task difficulty are big, and manual work easily regards tired
The problem of labor, easy maloperation, the present invention provide a kind of automobile engine cylinder-body sequence number vision automatic recognition method, and this method carries
The high stability and high efficiency, the various defects for overcoming artificial handwork of operation, has saved operating personnel's quantity.
Technical solution is as follows:
A kind of automobile engine cylinder-body sequence number vision automatic recognition method, step are as follows:
S1, collection automobile engine cylinder-body serial number image;
S2, train generation digital sort device according to sample Selection Model;
S3, read in all model images, carries out binary conversion treatment to image, profile is retrieved from bianry image, return to inspection
The profile number that rope arrives;
S4, read in the automobile engine cylinder-body serial number image for needing to identify, binary conversion treatment is carried out to image, from two-value
Profile is retrieved in image, returns to the profile number retrieved;
S5, circulation automobile engine cylinder-body serial number image profile information, are detected in the circulating cycle, cutting numeral/letter wheel
Exterior feature, in the same size with model image, the image after cutting subtracts each other with model image respective pixel point value, and it is whole to calculate gained picture
The quadratic sum of the pixel point value of picture, one group of corresponding model image of the quadratic sum minimum is numeral/word to be identified
It is female.
Further, camera is put on fixed distance, height shelf, is adjusted to fixed light, shooting automobile hair
Motivation cylinder body sequence number picture, distance, height when ensureing to shoot each automobile engine cylinder-body sequence number, daylighting are consistent.
Further, the sample uses general numeral sample storehouse or the specialized training made for application scenarios
Sample.
Further, all image template data are read in from specified folder using cvLoadImage, used
The processing of cvThreshold advances gray level image, then carries out threshold operation and obtains bianry image, threshold value is 100, is used
CvFindContours retrieves profile from bianry image.
Further, the automobile engine cylinder-body serial number image for needing to identify is read in using cvLoadImage, is used
The processing of cvThreshold advances gray level image, then carries out threshold operation and obtains bianry image, threshold value is 98, is used
CvFindContours retrieves profile from bianry image.
Further, each numeral of acquisition/alphabetical profile top left co-ordinate and wide high scope.
Further, step S5 is realized using OPenCV functions cvMatchShapes.
Further, after having identified all numeral/letters, a two-dimensional array is defined, it is one-dimensional to be used for storing what is identified
Numeral/letter, two dimension are used for storing the digital abscissa, are resequenced according to abscissa, list the data identified.
The beneficial effects of the invention are as follows:
Automobile engine cylinder-body sequence number vision automatic recognition method of the present invention improve operation stability and
High efficiency, the various defects for overcoming artificial handwork, have saved operating personnel's quantity.
Brief description of the drawings
Fig. 1 is to play the sequence number schematic diagram carved in the embodiment of the present invention 3 on engine cylinder body;
Fig. 2 is digital 0 schematic diagram of sample in the embodiment of the present invention 3;
Fig. 3 is digital 2 schematic diagrames of sample in the embodiment of the present invention 3;
Fig. 4 is the profile information figure of picture to be identified in the embodiment of the present invention 3;
Fig. 5 is the schematic diagram after picture cutting to be identified in the embodiment of the present invention 3;
Fig. 6 is the displaying figure after numeral rearrangement in the embodiment of the present invention 3.
Embodiment
Embodiment 1
A kind of automobile engine cylinder-body sequence number vision automatic recognition method, step are as follows:
S1, collection automobile engine cylinder-body serial number image;
S2, train generation digital sort device according to sample Selection Model;
S3, read in all model images, carries out binary conversion treatment to image, profile is retrieved from bianry image, return to inspection
The profile number that rope arrives;
S4, read in the automobile engine cylinder-body serial number image for needing to identify, binary conversion treatment is carried out to image, from two-value
Profile is retrieved in image, returns to the profile number retrieved;
S5, circulation automobile engine cylinder-body serial number image profile information, are detected in the circulating cycle, cutting numeral/letter wheel
Exterior feature, in the same size with model image, the image after cutting subtracts each other with model image respective pixel point value, and it is whole to calculate gained picture
The quadratic sum of the pixel point value of picture, one group of corresponding model image of the quadratic sum minimum is numeral/word to be identified
It is female.
Embodiment 2
A kind of automobile engine cylinder-body sequence number vision automatic recognition method, step are as follows:
S1, put camera on fixed distance, height shelf, is adjusted to fixed light, shoots automotive engine cylinder
Body sequence number picture, distance, height when ensureing to shoot each automobile engine cylinder-body sequence number, daylighting are consistent;
S2, train generation digital sort device according to sample Selection Model, and the sample uses general numeral sample storehouse;
S3, using cvLoadImage read in all image template data from specified folder, uses cvThreshold first
Into property gray level image processing, then carry out threshold operation obtain bianry image, threshold value is 100, using cvFindContours from
Profile is retrieved in bianry image, returns to the profile number retrieved;
S4, read in the automobile engine cylinder-body serial number image for needing to identify using cvLoadImage, uses
The processing of cvThreshold advances gray level image, then carries out threshold operation and obtains bianry image, threshold value is 98, is used
CvFindContours retrieves profile from bianry image, returns to the profile number retrieved;
S5, circulation automobile engine cylinder-body serial number image profile information, detect in the circulating cycle, acquisition it is each numeral/
Alphabetical profile top left co-ordinate and wide high scope, cutting numeral/alphabetical profile, figure cutting after in the same size with model image
As subtracting each other with model image respective pixel point value, the quadratic sum of the pixel point value of the calculating gained whole picture of picture, described square
It is numeral/letter to be identified with one group of minimum corresponding model image.Step S5 uses OPenCV functions
CvMatchShapes is realized.
After having identified all numeral/letters, a two-dimensional array is defined, it is one-dimensional to be used for storing the numeral/letter identified,
Two dimension is used for storing the digital abscissa, is resequenced according to abscissa, lists the data identified.
Embodiment 3
First, engine cylinder body sequence number vision-based detection:
The present embodiment mainly uses OpenCV numeral recognition detection engine cylinder body sequence numbers, comprises the steps of:
1. obtain picture;
2. collect digital picture template;
3. gone out using network analysis digital in picture.
2nd, the course of work:
Program is using Microsoft Visual Studio 2012 and OpenCV3.0 in 7-64 Ultimates of Windows
Develop and complete under system.And tested under 7-64 systems of Windows available.
Picture is to need to put camera on fixed distance, height shelf first, is adjusted to fixed light, shoots into
After picture, then numeral in picture is identified using the system.Have to early period to ensure distance, height, daylighting all it is unified can, send out
It is as shown in Figure 1 that the sequence number carved is beaten on motivation cylinder body.
Technical requirements and parameter:、
1. preliminary preparation:Official website (https://opencv.org/) latest edition OpenCV is downloaded, the system uses
3.0 version;
Configuration surroundings variable:In native en variable-system variable-path back addition;D:\Program Files\
OpenCV3.0\opencv\build\x86\vc12\bin;D:\Program Files\OpenCV3.0\opencv\build\
x64\vc12\bin
This 64 for translation and compiling environment be win32 compiling or win64 compiling
2. prepare template:According to sample, Selection Model training produces digital sort device.Here sample can be general
Numeral sample storehouse (such as NIST) or the specialized training sample for being directed to application scenarios and making.The former is excellent in generalization,
In accuracy rate, both are combined by force by the latter, i.e., are made an amendment on the basis of general digital storehouse.The sample number extracted according to application scenarios
As described in Figure 2, sample numeral 2 is as shown in Figure 3 for word 0.
3. analyze engine cylinder body sequence number
(1) template is read in, and is read in all image template data from specified folder using cvLoadImage, is used
CvThreshold binary images, i.e., carry out threshold operation to gray level image and obtain bianry image, threshold value is 100, is used
CvFindContours retrieves profile from bianry image, and returns to the number of the profile detected, only needs retrieval here most
The profile of outside.
(2) sequence number picture is handled:The same picture for needing to identify that read in using cvLoadImage is read with gray level image
Take;Using cvThreshold binary images, threshold value is 98, and profile is retrieved from bianry image using cvFindContours,
And the number of the profile detected is returned, need exist for retrieving all profiles, the profile information got is as shown in Figure 4.
(3) sequence number profile is handled:Profile information circulates, and detects in the circulating cycle, the digital profile upper left corner each obtained
Coordinate and wide high scope, as shown in figure 5, the digital profile of cutting chooses size of the size as template, then allow needs
The figure matched somebody with somebody subtracts each other with template respectively, and two picture respective pixel point values subtract each other, and then seeks the pixel for the whole picture for returning to picture
Point value obtains quadratic sum, and the quadratic sum minimum of picture is returned when which template matches and then can be obtained by as a result, using
OPenCV functions cvMatchShapes is realized.
(4) numeral arranges:A two-dimensional array is defined, one-dimensional to be used for storing the numeral identified, two dimension is used for storing this
The abscissa of numeral, contrast can find segmentation figure as when to search digital profile be not in order, it is necessary to further according to horizontal stroke
Coordinate is resequenced, and lists the data identified, as shown in Figure 6.
This scheme successfully solves the problems, such as that automobile engine cylinder-body sequence number identifies storage, and people can only be relied in the past by solving
Work industry, operation craftization, task difficulty are big, the easy visual fatigue of manual work, easy maloperation the problems such as;Improve transaction capabilities
Stability, high efficiency, overcome the various defects of artificial handwork, have saved operating personnel's quantity.
Embodiment 4
The present invention is used for other suitable for identifying that the product ID beaten and carved on automobile engine cylinder-body surface is also expansible
Part plays the identification for symbol sequence number of carving characters.
The foregoing is only a preferred embodiment of the present invention, but protection scope of the present invention be not limited thereto,
Any one skilled in the art in the technical scope of present disclosure, technique according to the invention scheme and its
Inventive concept is subject to equivalent substitution or change, should be covered by the protection scope of the present invention.
Claims (8)
- A kind of 1. automobile engine cylinder-body sequence number vision automatic recognition method, it is characterised in that step is as follows:S1, collection automobile engine cylinder-body serial number image;S2, train generation digital sort device according to sample Selection Model;S3, read in all model images, carries out binary conversion treatment to image, profile is retrieved from bianry image, return retrieves Profile number;S4, read in the automobile engine cylinder-body serial number image for needing to identify, binary conversion treatment is carried out to image, from bianry image Middle retrieval profile, returns to the profile number retrieved;S5, circulation automobile engine cylinder-body serial number image profile information, are detected in the circulating cycle, cutting numeral/alphabetical profile, with Model image is in the same size, and the image after cutting subtracts each other with model image respective pixel point value, calculates the gained whole picture of picture Pixel point value quadratic sum, one group of corresponding model image of the quadratic sum minimum is numeral/letter to be identified.
- 2. the automobile engine cylinder-body sequence number vision automatic recognition method as described in right will go 1, it is characterised in that in fixation Distance, put camera on height shelf, be adjusted to fixed light, shoot automobile engine cylinder-body sequence number picture, ensure Distance, height when shooting each automobile engine cylinder-body sequence number, daylighting are consistent.
- 3. the automobile engine cylinder-body sequence number vision automatic recognition method as described in right will go 1, it is characterised in that the sample This is using general numeral sample storehouse or the specialized training sample made for application scenarios.
- 4. automobile engine cylinder-body sequence number vision automatic recognition method as claimed in claim 1, it is characterised in that use CvLoadImage reads in all image template data from specified folder, at cvThreshold advance gray level images Reason, then carries out threshold operation and obtains bianry image, threshold value is 100, is retrieved using cvFindContours from bianry image Profile.
- 5. automobile engine cylinder-body sequence number vision automatic recognition method as claimed in claim 1, it is characterised in that use CvLoadImage reads in the automobile engine cylinder-body serial number image for needing to identify, uses cvThreshold advance gray-scale maps As processing, then carry out threshold operation and obtain bianry image, threshold value is 98, is examined using cvFindContours from bianry image Cable pulley is wide.
- 6. automobile engine cylinder-body sequence number vision automatic recognition method as claimed in claim 1, it is characterised in that acquisition Each numeral/alphabetical profile top left co-ordinate and wide high scope.
- 7. automobile engine cylinder-body sequence number vision automatic recognition method as claimed in claim 1, it is characterised in that step S5 Realized using OPenCV functions cvMatchShapes.
- 8. automobile engine cylinder-body sequence number vision automatic recognition method as claimed in claim 1, it is characterised in that identified After all numeral/letters, a two-dimensional array is defined, one-dimensional to be used for storing the numeral/letter identified, two dimension is used for storing this The abscissa of numeral, resequences according to abscissa, lists the data identified.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711408666.0A CN108009545A (en) | 2017-12-22 | 2017-12-22 | A kind of automobile engine cylinder-body sequence number vision automatic recognition method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711408666.0A CN108009545A (en) | 2017-12-22 | 2017-12-22 | A kind of automobile engine cylinder-body sequence number vision automatic recognition method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108009545A true CN108009545A (en) | 2018-05-08 |
Family
ID=62060697
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711408666.0A Pending CN108009545A (en) | 2017-12-22 | 2017-12-22 | A kind of automobile engine cylinder-body sequence number vision automatic recognition method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108009545A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1600889A1 (en) * | 2004-05-21 | 2005-11-30 | Samsung Electronics Co., Ltd. | Apparatus and method for extracting character(s) from image |
CN102364496A (en) * | 2011-11-24 | 2012-02-29 | 无锡慧眼电子科技有限公司 | Method and system for identifying automobile license plates automatically based on image analysis |
CN105809166A (en) * | 2016-03-04 | 2016-07-27 | 深圳市佳信捷技术股份有限公司 | Vehicle license plate recognition method, device and system |
CN106709530A (en) * | 2017-01-17 | 2017-05-24 | 中国科学院上海高等研究院 | License plate recognition method based on video |
-
2017
- 2017-12-22 CN CN201711408666.0A patent/CN108009545A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1600889A1 (en) * | 2004-05-21 | 2005-11-30 | Samsung Electronics Co., Ltd. | Apparatus and method for extracting character(s) from image |
CN102364496A (en) * | 2011-11-24 | 2012-02-29 | 无锡慧眼电子科技有限公司 | Method and system for identifying automobile license plates automatically based on image analysis |
CN105809166A (en) * | 2016-03-04 | 2016-07-27 | 深圳市佳信捷技术股份有限公司 | Vehicle license plate recognition method, device and system |
CN106709530A (en) * | 2017-01-17 | 2017-05-24 | 中国科学院上海高等研究院 | License plate recognition method based on video |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109087274B (en) | Electronic device defect detection method and device based on multi-dimensional fusion and semantic segmentation | |
CN109165538B (en) | Bar code detection method and device based on deep neural network | |
CN102982323B (en) | Gait recognition method fast | |
US20190188451A1 (en) | Lightweight 3D Vision Camera with Intelligent Segmentation Engine for Machine Vision and Auto Identification | |
CN110348263A (en) | A kind of two-dimensional random code image recognition and extracting method based on image recognition | |
CN110363803B (en) | Target detection method and system combining depth map slice and neural network | |
CN108108735A (en) | A kind of Automobile trade mark automatic identifying method | |
CN108009538A (en) | A kind of automobile engine cylinder-body sequence number intelligent identification Method | |
CN111178252A (en) | Multi-feature fusion identity recognition method | |
CN110751500B (en) | Processing method and device for sharing pictures, computer equipment and storage medium | |
CN104951440B (en) | Image processing method and electronic equipment | |
CN105574161A (en) | Trademark graph element identification method, device and system | |
CN110570442A (en) | Contour detection method under complex background, terminal device and storage medium | |
CN105389581A (en) | Germinated rice germ integrity intelligent identification system and identification method thereof | |
CN112149690A (en) | Tracing method and tracing system based on biological image feature recognition | |
CN109191461A (en) | A kind of Countryside Egg recognition methods and identification device based on machine vision technique | |
CN112215861A (en) | Football detection method and device, computer readable storage medium and robot | |
CN108121806A (en) | One kind is based on the matched image search method of local feature and system | |
CN111178405A (en) | Similar object identification method fusing multiple neural networks | |
CN114119695A (en) | Image annotation method and device and electronic equipment | |
CN111666813B (en) | Subcutaneous sweat gland extraction method of three-dimensional convolutional neural network based on non-local information | |
CN115661810A (en) | Security check CT target object identification method and device | |
CN112927194A (en) | Automatic checking method and system for design drawing and real object | |
CN110751732A (en) | Method for converting 2D image into 3D image | |
CN108009545A (en) | A kind of automobile engine cylinder-body sequence number vision automatic recognition method |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180508 |