CN105095851B - A kind of position of steel coil recognition methods - Google Patents
A kind of position of steel coil recognition methods Download PDFInfo
- Publication number
- CN105095851B CN105095851B CN201510095799.1A CN201510095799A CN105095851B CN 105095851 B CN105095851 B CN 105095851B CN 201510095799 A CN201510095799 A CN 201510095799A CN 105095851 B CN105095851 B CN 105095851B
- Authority
- CN
- China
- Prior art keywords
- picture
- coil
- strip
- circle
- white point
- 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)
Abstract
The invention belongs to the cargo field of image detection of cargo train, is specifically a kind of recognition methods of position of steel coil, mainly includes the following steps：Step 1, the top view of the railway carriage of acquisition, and it is translated into gray scale picture；Step 2, judges whether picture has strong light to disturb；Step 3, the identification of compartment wall position；Step 4, judges interference picture；Step 5, removes the miscellaneous point in picture；Step 6, position of steel coil identification.The present invention uses image recognition technology, image recognition is carried out by the vertical view picture of the railway carriage photographed to camera, it can rapidly identify the coil of strip in railway carriage, recognition speed is very fast, accuracy is high, reduces time when position of steel coil is checked during cargo train transport coil of strip and manpower consumption.
Description
Technical field
The invention belongs to the cargo field of image detection of cargo train, is specifically a kind of recognition methods of position of steel coil.
Background technology
When train steel coil transportation, the position of coil of strip should be ensured that in the middle position in compartment in compartment, if coil of strip
It has been displaced on the side in compartment, may have caused damage to coil of strip during train vibration influence.Therefore, on the way of steel coil transportation
In, the position for detecting coil of strip is a critically important process.
With the high speed development of railway construction in China, goods station's vehicles while passing is continuously increased, and the train disengaging time is short, single
Cargo difficulty and great work intensity are manually detected, while can also grow various security risks, therefore ensures safety of railway traffic,
The importance for reducing the working strength of Cargo Inspection operator on duty just gradually highlights.
The detection of train cargo mainly shoots each railway carriage by the high-definition camera on platform at present
The photo in left side, right side and top, photo is transmitted at train freight detection, by manually observing every photo, is judged
Compartment cargo is with the presence or absence of abnormal in photo.Long the time required to this detection method, consumption manpower is big.
Current information-based fast-developing, mode identification technology continues to develop, by carrying out image procossing to the photo of shooting,
The method that the method for detection compartment cargo exception ought to substitute the artificial judgment exception taken time and effort.
The content of the invention
The purpose of the present invention is to overcome the above shortcomings and to provide a kind of recognition methods of position of steel coil, reduces shipping fire
Time during inspection position of steel coil and manpower consumption during car steel coil transportation.
To realize above-mentioned technical purpose, scheme provided by the invention is：A kind of recognition methods of position of steel coil, including it is as follows
Step.
Step 1, the top view of the railway carriage of acquisition, and it is translated into gray scale picture.
Step 2, judges whether picture has strong light to disturb, the gray scale picture in the vertical direction that step 1 is converted
It is equally divided into、、Three parts,WithTwo parts take the frequency sampling of a point according to horizontal, every ten points of ordinate
Obtain、The number of two parts point isWith, then sum respectively、Total gray scale of fractional-sample point isWith, then calculateWithThe total average gray value of two parts；
The frequency sampling that part also takes a point according to horizontal, every ten points of ordinate calculates average gray value,
Ask againAverage gray value with、The ratio of part average gray valueIf the ratio is more than
1.35, then judge that this center picture part has strong light to disturb.
Step 3, compartment wall position identification, is obtained with step 2WithThe total average gray value of two partsWithPartial average gray valueCalculate the average gray value of picture, according to
Binary-state threshold is calculated：If no strong light interference,；If
There is strong light to disturb,, with this binary-state thresholdTo picture binaryzation,
Obtained two-value picture is traveled through, calculates the white point number of each column in two-value picture, judge the most point of white point is classified as compartment wall
Position.
Step 4, judges interference picture, if center picture partial shape is set close to the bulk white point region of rectangle, its width
For, and＞ 500, its length are set to, and＞ 1300, calculates two-value picture white point number and is more thanRow company
Continuous columnsIf, then it is interference picture to judge time picture, i.e., does not have coil of strip in picture, calculatingWhen, row
Except in pictureWithEach 300 points of two parts, prevent interference caused by the white point of compartment wall up and down after picture binaryzation.
Step 5, removes the miscellaneous point in picture, if the profile girth in the region where miscellaneous point is set to, and＜ 200, is used
Suzuki85 algorithms take the profile of picture, then the boundary rectangle of profile of the contouring length less than 200, and the point in rectangle is
It is miscellaneous, miscellaneous point is removed.
Step 6, position of steel coil identification, first judges whether there is circular ring shape coil of strip in picture, if so, then exporting position of steel coil；
If no, then judging whether there is rectangle coil of strip in picture, if so, output position of steel coil；If no, judge there is no steel in picture
Volume.
Moreover, the circular ring shape coil of strip recognition methods in the step 6 is, by Hall circle transformation, take out in picture
Circle, if the center of circle for the circle that Hall circle transformation obtains is, radius is, the following either condition of satisfactory foot of Hall circle transformation generation
It is as effectively round：
A) in bianry image, distance is calculatedLength isThe number of white point be, radius isCircle week
It is a length ofIf, then it is effectively circle；
B) in bianry image, distance is calculatedLength isThe number of white point be, radius isCircle Zhou ChangweiIf, then it is effectively circle；
C) in bianry image, distance is calculatedLength isThe number of white point be, radius isCircle Zhou ChangweiIf, then it is effectively circle；
Wherein,。
Moreover, the rectangle coil of strip recognition methods in the step 6 is, in binary image, it is most to calculate white point number
One row white point number, be set to, the height of each rectangle coil of strip is set to L,
If a), then there is no coil of strip in picture；
If b)And, then；
If c), then；
Then the white point number of picture each column is calculated from left to right on bianry image,
If a)And, then continuous white point number is judgedAndRow
NumberWhether reach,If reaching, there is a coil of strip herein；
If b), then continuous white point number is judgedRow numberWhether reach,If reaching, there are two coil of strips herein；
Calculating continuous columnsWhen, when the number for the row for not meeting the condition of continuity is more thanWhen,, judge
It is discontinuous at this time.
The beneficial effects of the present invention are：The present invention uses image recognition technology, passes through the train photographed to camera
The vertical view picture in compartment carries out image recognition, can rapidly identify the coil of strip in railway carriage, recognition speed is very fast, accurately
Degree is high, reduces time when position of steel coil is checked during cargo train transport coil of strip and manpower consumption.
Brief description of the drawings
Fig. 1 is the flow chart of the present invention.
Fig. 2 is the original image for having in the compartment for needing to identify circular ring shape coil of strip.
Fig. 3 is the gray scale picture for the picture for having circular ring shape coil of strip.
Fig. 4 is the picture after the binaryzation for the picture for having circular ring shape coil of strip.
Fig. 5 is the position picture of the circular ring shape coil of strip detected.
Fig. 6 is the original image for having in the compartment for needing to identify rectangle coil of strip.
Fig. 7 is the gray scale picture for the picture for having rectangle coil of strip.
Fig. 8 is the picture after the binaryzation for the picture for having rectangle coil of strip.
Fig. 9 is the position picture of the rectangle coil of strip detected.
Embodiment
The invention will be further described with reference to the accompanying drawings and embodiments.
The present embodiment provides a kind of recognition methods of position of steel coil, as shown in Figure 1, including the following steps.
1st, can by the camera that is installed on above railway, the top view of the railway carriage of acquisition, such as Fig. 2 and Fig. 6, and by its
Gray scale picture is converted into, such as Fig. 3 and Fig. 7.
2nd, judge whether picture has strong light to disturb.
The strong light interference of picture is to have a strong light in the middle part of picture caused by the flash lamp flicker of camera in shooting process
Band, in order to avoid erroneous judgement caused by euphotic zone, first judges whether picture has strong light to disturb.
The gray scale picture in the vertical direction that conversion obtains is equally divided into、、Three parts,WithTwo
Divide and take the frequency sampling of a point to obtain according to horizontal, every ten points of ordinate、The number of two parts point isWith, connect
And sum respectively、Total gray scale of fractional-sample point isWith, then calculateWithThe total average gray of two parts
Value.Also the frequency of a point is taken according to horizontal, every ten points of ordinate
Sampling calculates average gray value, askAverage gray value with、The ratio of part average gray value.And the highlights of obtained picture is shot in picturePart, is detected by analyzing to calculate
Picture, obtains the picture of strong light interferenceIt is all higher than, and the picture without the interference of strong lightRespectively less than,
Therefore takeCut off valueIf, then this center picture part there is strong light to disturb.
3rd, compartment wall position identifies.
With obtained above、Part average gray valueWithAverage gray valueCalculate being averaged for picture
Gray value, according toBinary-state threshold is calculated：If no strong light interference,, if there is strong light to disturb,
, with this binary-state thresholdTo picture binaryzation.Obtained two-value picture is traveled through, such as Fig. 4 and Fig. 8, is calculated every in two-value picture
The white point number of row, the row of the most point of white point are the position of compartment wall.
4th, interference picture is judged.
Interference picture is mainly the vertical view picture that tank car or passenger car are loaded in compartment.Tank car and car top view two
After value, center picture part has shape close to the bulk white point region of rectangle, judges the white point region, you can judge
Disturb picture.The binaryzation picture of analysis tank car and car obtains, and the width of bulk white point rectangle is all higher than 500, is set to, it is long
Degree is all higher than 1300, is set to.Therefore, calculating two-value picture white point number, which is more than, isRow continuous columnsIf, this picture is the picture of tank car or car, that is, disturbs picture, do not have coil of strip in picture.CalculatingWhen, row
Except in pictureWithEach 300 points of two parts, prevent interference caused by the white point of compartment wall up and down after picture binaryzation.
5th, the miscellaneous point in picture is removed.
Then the miscellaneous point of picture is removed.Miscellaneous point is the point or point set of dispersed distribution in picture after binaryzation, passes through analysis two
Value picture, the girth in the region where miscellaneous point are respectively less than 200, are set to.The profile of picture is taken with Suzuki85 algorithms, then
Contouring length is less thanProfile boundary rectangle, the point in rectangle is miscellaneous point, and miscellaneous point is removed.
6th, position of steel coil identifies.
Coil of strip has two states, horizontal positioned and vertical placement in railway carriage.When horizontal positioned, coil of strip shape approaches
For annulus, when placing vertically, position of steel coil is close to rectangle.During image recognition, first judge whether there is circular ring shape coil of strip in picture,
If so, position of steel coil is then exported, such as Fig. 5；If no, then judging whether there is rectangle coil of strip in picture, if so, output coil of strip position
Put, such as Fig. 9；If no, there is no coil of strip in picture.
6.1st, circular ring shape coil of strip identifies.
By Hall circle transformation, the circle in picture is taken out.Hall circle transformation can produce be not coil of strip interference circle.If Hall
The center of circle for the circle that circle transformation obtains is, radius is, analyze picture feature and obtain, justify if interference, then distance in bianry imageLength isWhite point number not less than circle girth。
So judge whether circle is interference circle by the following conditions.
A) in bianry image, distance is calculatedLength isThe number of white point be, radius isCircle week
It is a length ofIf, then it is effectively circle.
But since the circle that Hall circle transformation obtains may be the inscribed or circumcircle of circular coil of strip, use is above-mentioned
Judgement may filter out effective circle.
So in addition Rule of judgment will add following condition, fall effectively round situation to exclude a condition filters.
B) in bianry image, distance is calculatedLength isThe number of white point be, radius isCircle Zhou ChangweiIf, then it is effectively circle.
C) in bianry image, distance is calculatedLength isThe number of white point be, radius isCircle Zhou ChangweiIf, then it is effectively circle.
Wherein,, it is to analyze the length value that pictorial information obtains.
It is effectively round as long as above-mentioned a, b, c either condition of satisfactory foot that Hall circle transformation produces.
Finally, the effective circle detected is marked, the position of steel coil as detected.If the identification of circular ring shape coil of strip is not known
Circle is clipped to, then performs rectangle coil of strip identification process.
6.2nd, rectangle coil of strip identifies.
In binary image, to exclude the interference of compartment wall up and down, exclude in pictureWithEach 300 points of two parts,
Then the white point number of the largest number of row of white point is calculated, is set to, the height of each rectangle coil of strip is set to L.Pass through analysis chart
Picture, obtains following character.
If a), there is no coil of strip in picture.Because the white point sum of each row is unsatisfactory for coil of strip in picture
Minimum constructive height.
If b)And,。
If c),。
Then the white point number of picture each column is calculated from left to right on bianry imageIfAnd,
Then judge continuous white point numberAndRow numberWhether reachIf reaching, herein
There is a coil of strip；If, then continuous white point number is judgedRow numberWhether reach,
If reaching, there are two coil of strips herein.CalculatingWhen, do not connect because being likely to occur in binary image inside the white point of coil of strip
Continuous situation, therefore when calculating continuation column, only when the number for the row for not meeting the condition of continuity is more thanWhen, just say at this time not
Continuously.Wherein, analysis image information obtains,。
Finally, the rectangle steel label detected is gone out, the position of steel coil as detected.
The above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, without departing from the principle of the present invention, some improvement or deformation can also be made, these are improved or deformation also should
It is considered as protection scope of the present invention.
Claims (3)
1. a kind of recognition methods of position of steel coil, includes the following steps：
Step 1, the top view of the railway carriage of acquisition, and it is translated into gray scale picture；
Step 2, judges whether picture has strong light to disturb, and the gray scale picture in the vertical direction that step 1 converts is averaged
It is divided into、、Three parts,WithTwo parts take the frequency sampling of a point to obtain according to horizontal, every ten points of ordinate、The number of two parts point isWith, then sum respectively、Total gray scale of fractional-sample point isWith,
Then calculateWithThe total average gray value of two parts；Part
Also the frequency sampling for taking a point according to horizontal, every ten points of ordinate calculates average gray value, then askAverage gray value with、The ratio of part average gray valueIf the ratio is more than 1.35,
Judge that this center picture part has strong light to disturb；
Step 3, compartment wall position identification, is obtained with step 2WithThe total average gray value of two partsWithPortion
The average gray value dividedCalculate the average gray value of picture, according toCalculate
Obtain binary-state threshold：If no strong light interference,；It is strong if having
Light disturbs,, with this binary-state thresholdTo picture binaryzation, traversal
Obtained two-value picture, calculates the white point number of each column in two-value picture, judges the position for being classified as compartment wall of the most point of white point
Put；
Step 4, judges interference picture, if center picture partial shape is set to close to the bulk white point region of rectangle, its width
, and＞ 500, its length are set to, and＞ 1300, calculates two-value picture white point number and is more thanRow it is continuous
ColumnsIf, then it is interference picture to judge this picture, i.e., does not have coil of strip in picture, calculatingWhen, exclude figure
In pieceWithEach 300 points of two parts, prevent interference caused by the white point of compartment wall up and down after picture binaryzation；
Step 5, removes the miscellaneous point in picture, if the profile girth in the region where miscellaneous point is set to, and＜ 200, is used
Suzuki85 algorithms take the profile of picture, then the boundary rectangle of profile of the contouring length less than 200, and the point in rectangle is
It is miscellaneous, miscellaneous point is removed；
Step 6, position of steel coil identification, first judges whether there is circular ring shape coil of strip in picture, if so, then exporting position of steel coil；If not yet
Have, then judge whether there is rectangle coil of strip in picture, if so, output position of steel coil；If no, judge there is no coil of strip in picture.
A kind of 2. recognition methods of position of steel coil according to claim 1, it is characterised in that：Annulus in the step 6
Shape coil of strip recognition methods is, by Hall circle transformation, the circle in picture to be taken out, if the center of circle for the circle that Hall circle transformation obtains is
, radius is, the following either condition of satisfactory foot that Hall circle transformation produces is effective circle,
In bianry image, distance is calculatedLength isThe number of white point be, radius isCircle Zhou Changwei,
If, then it is effectively circle；
In bianry image, distance is calculatedLength isThe number of white point be, radius isCircle
Zhou ChangweiIf, then it is effectively circle；
In bianry image, distance is calculatedLength isThe number of white point be, radius isCircle
Zhou ChangweiIf, then it is effectively circle；
Wherein,。
A kind of 3. recognition methods of position of steel coil according to claim 1, it is characterised in that：Rectangle in the step 6
Coil of strip recognition methods is, in binary image, calculates the white point number of the largest number of row of white point, is set to, each rectangle
The height of coil of strip is set to L,
If, then there is no coil of strip in picture；
IfAnd, then；
If, then；
Then the white point number of picture each column is calculated from left to right on bianry image,
IfAnd, then continuous white point number is judgedAndRow numberIt is
It is no to reach,If reaching, there is a coil of strip herein；
If, then continuous white point number is judgedRow numberWhether reach,If reaching, there are two coil of strips herein；
Calculating continuous columnsWhen, when the number for the row for not meeting the condition of continuity is more thanWhen,, judge at this time not
Continuously.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510095799.1A CN105095851B (en) | 2015-03-04 | 2015-03-04 | A kind of position of steel coil recognition methods |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510095799.1A CN105095851B (en) | 2015-03-04 | 2015-03-04 | A kind of position of steel coil recognition methods |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105095851A CN105095851A (en) | 2015-11-25 |
| CN105095851B true CN105095851B (en) | 2018-04-24 |
Family
ID=54576241
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510095799.1A Active CN105095851B (en) | 2015-03-04 | 2015-03-04 | A kind of position of steel coil recognition methods |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105095851B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105956619B (en) * | 2016-04-27 | 2019-05-24 | 浙江工业大学 | A kind of container lockhole coarse positioning and tracking |
| CN109635797B (en) * | 2018-12-01 | 2023-01-13 | 北京首钢自动化信息技术有限公司 | Steel coil sequence accurate positioning method based on multi-carrier identification technology |
| CN109976348A (en) * | 2019-04-11 | 2019-07-05 | 深圳市大富科技股份有限公司 | A kind of vehicle and its motion control method, equipment, storage medium |
| CN110113517A (en) * | 2019-05-27 | 2019-08-09 | 杭州亚美利嘉科技有限公司 | Exempt from the camera and self-power supply device of wiring |
| CN112132053A (en) * | 2020-09-24 | 2020-12-25 | 杭州海康威视数字技术股份有限公司 | Carriage identification method and device and computer storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101950354A (en) * | 2010-09-03 | 2011-01-19 | 西安大唐电信有限公司 | System and method for recognizing national railway wagon number information image |
| CN102768731A (en) * | 2012-06-29 | 2012-11-07 | 陕西省交通规划设计研究院 | Method and system for automatic positioning and identifying target based on high definition video images |
| CN103077387A (en) * | 2013-02-07 | 2013-05-01 | 东莞中国科学院云计算产业技术创新与育成中心 | Method for automatically detecting carriage of freight train in video |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102011077296B4 (en) * | 2011-06-09 | 2020-12-10 | Carl Zeiss Smt Gmbh | Method and device for determining the position of a first structure relative to a second structure or a part thereof |
-
2015
- 2015-03-04 CN CN201510095799.1A patent/CN105095851B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101950354A (en) * | 2010-09-03 | 2011-01-19 | 西安大唐电信有限公司 | System and method for recognizing national railway wagon number information image |
| CN102768731A (en) * | 2012-06-29 | 2012-11-07 | 陕西省交通规划设计研究院 | Method and system for automatic positioning and identifying target based on high definition video images |
| CN103077387A (en) * | 2013-02-07 | 2013-05-01 | 东莞中国科学院云计算产业技术创新与育成中心 | Method for automatically detecting carriage of freight train in video |
Non-Patent Citations (2)
| Title |
|---|
| Video events recognition by improved stochastic parsing based on extended stochastic context-free grammar representation;CAO Mao-yong 等;《Journal of Beijing Institute of Technology》;20130315;第22卷(第1期);第81-88 * |
| 钢卷空间位置识别技术的研究;张兰;《软件》;20140115;第35卷(第1期);第69-71页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105095851A (en) | 2015-11-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105095851B (en) | A kind of position of steel coil recognition methods | |
| CN103499585B (en) | Based on noncontinuity lithium battery film defect inspection method and the device thereof of machine vision | |
| CN106448180B (en) | A kind of major long tunnel traffic events real-time detection method and detection system | |
| CN105788142A (en) | Video image processing-based fire detection system and detection method | |
| CN109766746B (en) | Track foreign matter detection method for aerial video of unmanned aerial vehicle | |
| CN110532852A (en) | Subway station pedestrian's accident detection method based on deep learning | |
| CN107316010A (en) | A kind of method for recognizing preceding vehicle tail lights and judging its state | |
| KR102017870B1 (en) | Real-time line defect detection system | |
| CN103324930A (en) | License plate character segmentation method based on grey level histogram binaryzation | |
| CN105373782A (en) | Method of automatically recognizing hazardous chemical vehicle from image or video | |
| CN102999749A (en) | Intelligent safety belt regulation violation event detecting method based on face detection | |
| CN111080607B (en) | Rolling bearing oil slinging fault detection method based on image recognition | |
| CN110562292B (en) | Dynamic detection system for diameter of railway vehicle wheel set | |
| CN111626170A (en) | Image identification method for railway slope rockfall invasion limit detection | |
| CN107358270B (en) | Water seepage detection method and device for tunnel wall | |
| CN113657265B (en) | Vehicle distance detection method, system, equipment and medium | |
| CN108764115A (en) | A kind of truck danger based reminding method | |
| CN111738204A (en) | Vehicle tire stability analysis system and method based on Internet of things | |
| US11836994B2 (en) | Methods and apparatus for automatic collection of under-represented data for improving a training of a machine learning model | |
| CN111125903A (en) | Method for calculating evacuation reliability of fire personnel in subway tunnel train | |
| CN111091551A (en) | Method for detecting loss fault of brake beam strut opening pin of railway wagon | |
| CN113610014A (en) | System and method for detecting freight vehicle with shielding number plate exceeding limit | |
| CN113657305B (en) | Video-based intelligent detection method for black smoke vehicle and ringeman blackness level | |
| JP6307800B2 (en) | Rail inspection apparatus and inspection method | |
| CN100585618C (en) | A vehicle detection method based on color video |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 430070 Hubei province Wuhan city Hongshan District Hongshan town Wenzhi Street No. 16 building 1-7 layer 1 Applicant after: Wuhan Lide Control Technology Co. Address before: 430070 Hubei province Wuhan city Hongshan District Hongshan town Wenzhi Street No. 16 building 1-7 layer 1 Applicant before: WUHAN LIDE SOFTWARE CO., LTD. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |