CN102024144A - Container number identification method - Google Patents
Container number identification method Download PDFInfo
- Publication number
- CN102024144A CN102024144A CN 201010554679 CN201010554679A CN102024144A CN 102024144 A CN102024144 A CN 102024144A CN 201010554679 CN201010554679 CN 201010554679 CN 201010554679 A CN201010554679 A CN 201010554679A CN 102024144 A CN102024144 A CN 102024144A
- Authority
- CN
- China
- Prior art keywords
- character
- image
- picture
- case number
- feature
- 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
Images
Landscapes
- Character Discrimination (AREA)
Abstract
The invention discloses a container number identification method which comprises the following steps of, firstly, after acquiring pictures of a container number, searching a frame of a container, calculating the inclination with the frame as a reference line, and intercepting pictures in the frame subjected to correction; secondly, carrying out graying and dynamic binarization on the intercepted pictures, and filtering noise interference; thirdly, scanning and segmenting container number character pictures which satisfy the conditions one by one through a rank scanning method; fourthly, accurately extracting 11-bit container number characters which satisfy the conditions by determining an equipment identifier; and fifthly, extracting character characteristics one by one to finish the identification process, and judging whether the container number is correct by combining a check bit. The invention improves the process of respectively carrying out location, segmentation and identification of the traditional container number identification algorithm, enhances the continuity of a whole system and also has the advantages of rapid identification speed and high accuracy.
Description
Technical field
The present invention relates to a kind of image recognition technology, be specifically related to a kind of container number automatic identification technology.
Background technology
Each container all has their unique case number (CN)s, these case number (CN)s are container accessing to the ports each time, the warehouse, all can be used to during customs, case number (CN) can appear on five faces of container (bottom surface does not have case number (CN)), and present stage majority harbour, the harbour, customs and warehouse all use the operator to remove to write down these case number (CN)s, are input to then in the computer, and this has not only wasted plenty of time and manpower but also meeting because operator's various carelessness cause misregistration.
Based on the problems referred to above, increasing people expects that reading character that machine can be quick and correct automatically and numeral are in order to increase work efficiency.Because the automatic identification technology of automotive license plate, it is applied in traffic department widely, the number-plate number that the detection that the department that help has to be needed does not day and night stop grabs by the video grasping system, thus significantly reduced human cost, accelerate work efficiency and improved the work accuracy rate.Thereby design the case number (CN) recognition system of corresponding container.
Yet the realization of automatic recognition system also is faced with many difficulties, because the case number (CN) of container does not have fixing font and size, the position of case number (CN) also can change along with the difference of chest, even in the same chest, the orientation of coplanar top box number can be not different with the position yet, and these all can bring certain difficulty to identification.
When discerning automatically, the initial container picture that has case number (CN) comes from the video grabbing device, and (relatively more commonly used is the camera that the automatic focusing performance is arranged, also can use video camera etc.), but the solar irradiation intensity on daytime, evening auto exposure system the exposure ability, even the reflecting rate of container finishing coat all can influence the last recognition result of system greatly, so in order to replace existing people manually to write down the case number (CN) mode, a is stable, accurately, the exploitation that can adapt to strong automatic character recognition system to surrounding environment extremely is necessary.
Existing container number automatic identification technology is generally all copied the comparatively ripe automatic Recognition of License Plate of present development, the identifying of case number (CN) is divided into " location of case number (CN), Character segmentation, feature extraction and character recognition ", wherein case number (CN) location and Character segmentation are thought the committed step in the case number (CN) automatic identification technology in the industry.But compare with the characters on license plate arrangement, case number (CN) is arranged varied, and its scrambling has determined to carry out with traditional license plate locating method the case number (CN) location.
Publication number is CN 101650780A, the patented claim (application number is 200910170227.X) that open day is on February 17th, 2010 discloses a kind of container number identification method, it may further comprise the steps: the extraction of container number: the picture that will obtain comprising container number cuts out useful case number (CN) through image processing techniques, and the case number (CN) that cuts out is passed through image standardized processing once more; Character recognition: obtain through the information after the image standardized processing, use nerual network technique and Euclidean distance method that the character after cutting is discerned.Its algorithm computation complexity of this recognition methods, recognition speed is slow, inefficiency.
Therefore, providing the container number automatic identifying method that a kind of algorithm is simple, recognition speed is fast is the problem that this area is needed solution badly.
Summary of the invention
The present invention is directed to problems such as the existing existing recognition efficiency of container number recognition technology is low, and provide a kind of container number automatic identifying method, case number (CN) location is effectively merged the accurate extraction that realizes case number (CN) with cutting apart, greatly reduce the complicacy of algorithm simultaneously, simultaneously on character recognition algorithm, done corresponding improvement, raised the efficiency.
In order to achieve the above object, the present invention adopts following technical scheme:
The container number automatic identifying method may further comprise the steps:
(1) image inclination is proofreaied and correct: after obtaining the case number (CN) picture, search set vanning frame is thought line of reference measuring and calculating degree of tilt with this frame, proofreaies and correct picture in the back intercepting frame;
(2) image pre-service: with truncated picture gray processing, dynamically binaryzation, filtering noise interference;
(3) dividing box sign character:, scan and be partitioned into " case number (CN) character " picture that satisfies condition one by one by the rank scanning method;
(4) determining apparatus identification code:, accurately extract 11 case number (CN) characters that satisfy condition by the determining apparatus identification code;
(5) case number (CN) character recognition and verification: extract character feature one by one, finish identifying, and judge in conjunction with case number (CN) verification rule whether recognition result is accurate.
In the preferred embodiment of the present invention, the image pre-service of described step (2) realizes by following steps:
(2.1) gray processing
Read the progress R of the pixel in the data bitmap district of the image of (1) processing suddenly, G, the B tristimulus values is converted into gray-scale map by the gradation conversion formula with picture:
Wherein Grey is the gradation of image value after changing, and R, G, B are respectively redness, green and the blue component of original color image;
(2.2) dynamic binaryzation
Add up the bitmap pixels gray average and the mean square deviation of gray processing, calculate the dynamic binary-state threshold of view picture bitmap by following formula:
Wherein
Be grey scale pixel value,
Be respectively and get average and mean square deviation computing;
After obtaining this threshold value, image is carried out the binaryzation process, if grey scale pixel value greater than this threshold value, then is set at 255 with this grey scale pixel value, otherwise is 0;
(3.3) filtering noise
Utilize door lock Metallic rod interference in the linear feature filtering image: after the binaryzation, by the Hough conversion, the thread of scan image, when the row of straight line surpass certain threshold value to the foreground point pixel count, think that this straight line is that Metallic rod is listed as to the border, when the spacing of two metal line during less than certain threshold value, assert that the line-to-line zone is the Metallic rod district, this area pixel gray-scale value is made as 0;
Adopt morphological image algorithm opening operation to desalinate such character gradually, reduce to disturb: its formula is as follows:
Wherein S represents to carry out the bianry image set of opening operation, and B represents to be used for carrying out the structural element of opening operation, the collection of pixels after the X representation element image process binaryzation,
Represent that respectively opening operation, burn into expand.
Further, the dividing box sign character of described step (3) is realized by following steps:
(3.1) by the case number (CN) picture of lining by line scan, the searching image prospect is counted, and when surpassing certain threshold value when counting, tentatively regards as character zone, and makes mark, and whether need to distinguish one's own profession in this process is the up-and-down boundary of character zone;
(3.2) scanning entire image is demarcated all row zones that comprises character;
(3.3) after the line scanning, to demarcating the row zone picture is listed as to scanning, when being listed as when image is counted above certain threshold value, character exists, and by edge determination, demarcates its border, left and right sides, simultaneously in conjunction with going regional this character of cutting.
Further, the determining apparatus identification code of described step (4) realizes by following steps:
(4.1) at the character picture of well cutting bottom three/part scanning straight line, if picture below straight line exists, then preliminary judgement is EIC equipment identification code " U ", otherwise thinks the non-EIC equipment identification code of character " U ";
(4.2) be on the character picture of " U " at preliminary judgement, flash trimming out-of-bounds three minutes highly on the statistics saltus step count, count in the statistics saltus step of character picture width equinox place, satisfy 4,4,2 if three saltus steps are counted, think that then character " U " exists.
Further, the case number (CN) character recognition and the checking procedure of described step (5) realize by following steps:
(5.1) extract 0 ~ 90 arabic numeral from cut apart good container code characters, totally 26 English alphabets of A ~ Z carries out size normalization with each character, and place normalization becomes the 42*28 pixel, does refinement again and handles, and makes template;
(5.2) each character is divided into equably six grids, add up character pixels number in each grid, form sextuple proper vector, draw bar line computation point of crossing number in the vertical orientation of character again, horizontal trisection posistion is drawn two horizontal lines and is calculated the point of crossing number, forms nine dimensional feature vectors altogether;
(5.3) 36 characters in the template are carried out feature extraction according to step (5.2), form the feature database of 36*9;
(5.4) extract the feature of character to be identified according to above-mentioned feature extracting method, English alphabet is separated with numeral discern, its identification principle is according to the method for Weighted distance template matches, and formula is as follows:
The feature of English alphabet to be identified and numeral is calculated Weighted distance with template base Chinese and English letter and number feature according to above-mentioned formula respectively, minimum with the distance of which template character, then just be this template character with character recognition to be identified;
(5.5) after identifying every character, use the case number (CN) check bit that identification case number (CN) is tested.
The present invention adopts special rank scanning method to carry out character locating to cut apart, and it extracts the efficient of improved the location, cutting apart by certain Rule of judgment with all " char " pictures that satisfy the case number (CN) character feature in the picture.
The present invention is by merging each algorithm mutually, guarantee that total system can effectively will meet the character extraction of case number (CN) character feature and identify, improve the process that traditional case number (CN) recognizer is located respectively, cuts apart, discerned, improved the continuity of total system.Recognition speed of the present invention is fast simultaneously, precision is high.
Description of drawings
Further specify the present invention below in conjunction with the drawings and specific embodiments.
Fig. 1 is the schematic flow sheet of the inventive method.
The former figure that Fig. 2 obtains for equipment.
Fig. 3 is the interior figure of Hough conversion intercepting.
Fig. 4 is image pretreating effect figure.
The character that Fig. 5 obtains for scanning method.
Fig. 6 is the design sketch of location character U.
Fig. 7 is a case number (CN) recognition result design sketch.
Embodiment
For technological means, creation characteristic that the present invention is realized, reach purpose and effect is easy to understand, below in conjunction with concrete diagram, further set forth the present invention.
Container number is formed by 11, wherein front three is the case main code, be capitalization English letter, the 4th is EIC equipment identification code-character " U ", six of backs are concrete case number (CN) numeral, last by rectangular shaped rim surround for the check bit numeral, it is obtained by the certain calculation combination by the top ten character.Each face of container all has case number (CN), but the case number (CN) irregular arrangement, and its pattern divides delegation, two row, four lines, row four kinds.The present invention is based on above container number characteristics and propose novel automatic identifying method.
Referring to Fig. 1, container number automatic identifying method provided by the invention may further comprise the steps:
(1) image inclination is proofreaied and correct
After obtaining the case number (CN) picture, grasp casing frame straight line by the Hough conversion, choose casing coboundary and right margin usually, and think that with this frame line of reference does slant correction, be that benchmark extends into prescribed level in the below left with two borders again, proofread and correct picture in the back intercepting frame.
(2) image pre-service
Will be by step (1) truncated picture gray processing, dynamically binaryzation, filtering noise interference.
Obtain 24 bitmaps that contain case number (CN) by camera at case tail or case side, it is carried out the image pre-service, comprising:
1, gray processing
Read the R of pixel in the data bitmap district of this bitmap, G, the B tristimulus values, (wherein each pixel is by three bytes store) are converted into gray-scale map by gradation conversion formula (formula as follows) with picture,
，
Wherein Grey is the gradation of image value after changing, and R, G, B are respectively redness, green and the blue component of original color image.
2, dynamic binaryzation
Add up the bitmap pixels gray average and the mean square deviation of gray processing, calculate the dynamic binary-state threshold of view picture bitmap by following formula:
Wherein
Be grey scale pixel value,
Be respectively and get average and mean square deviation computing.
After obtaining this threshold value, image is carried out the binaryzation process, if grey scale pixel value greater than this threshold value, then is set at 255 with this grey scale pixel value, otherwise is 0.This dynamic binaryzation algorithm can carry out threshold setting in conjunction with the picture characteristics, effectively keeps the character zone that comprises high brightness, reduces the interference of the non-character of low-light level simultaneously.
3, filtering noise
Observe a large amount of container number pictures, the case number (CN) picture that discovery is taken at the case tail generally all comprises chamber door lock Metallic rod, exist simultaneously the character of a large amount of mark case master information below case number (CN), these all cause interference to 11 case number (CN) code identifying, therefore need effective filtering.
Disturb for the door lock Metallic rod, utilize the in addition filtering of its linear feature.After the binaryzation, by the Hough conversion, the thread of scan image, when the row of straight line surpass certain threshold value (be generally picture altitude 60%) to the foreground point pixel count, think that this straight line is that Metallic rod is listed as to the border, when the spacing of two metal line during less than certain threshold value, assert that the line-to-line zone is the Metallic rod district, this area pixel gray-scale value is made as the 0(background dot).
Compare with 11 case number (CN) codes, the character that comprises case master information is usually individual much smaller, gesticulates also carefullyyer, so we adopt morphological image algorithm opening operation to desalinate such character gradually, reduce to disturb.Its formula is as follows:
Wherein S represents to carry out the bianry image set of opening operation, and B represents to be used for carrying out the structural element of opening operation, and X representation element image is through the collection of pixels after the binaryzation.
Represent that respectively opening operation, burn into expand.
(3) dividing box sign character
To scan and be partitioned into " case number (CN) character " picture that satisfies condition one by one by the pretreated image of step (2) by the rank scanning method.
Only have a few regions to keep higher signal noise ratio (snr) of image through above-mentioned pretreated case number (CN) picture, this is comprising the faint character of 11 case number (CN) characters, 4 heavy information characters (referring to " 42G1 " among Fig. 4) of case and mark case master information etc.
By the thought of horizontal projection method, by the case number (CN) picture of lining by line scan, the searching image prospect is counted, and when surpassing certain threshold value when counting, tentatively regards as character zone, and makes mark.Whether need to distinguish one's own profession in this process is the up-and-down boundary of character zone.The scanning entire image is demarcated all row zones that comprises character.
After the line scanning, to demarcating the row zone picture is listed as to scanning, in like manner, when being listed as when image is counted above certain threshold value, character exists, and by edge determination, demarcates its border, left and right sides, simultaneously in conjunction with going regional this character of cutting.
On threshold setting, transversely arranged character is with vertically arrangement character threshold value is different, and general case trailing character is laterally, and case side case number (CN) is vertically arranged.
(4) determining apparatus identification code
By the determining apparatus identification code, accurately extract 11 case number (CN) characters that satisfy condition.
In 11 case number (CN)s, wherein the 4th is EIC equipment identification code---character " U ", therefore, at accurate location character " U " but the position of 11 case number (CN)s of accurate localization.Because character " U " is obvious than other English, numerical character feature, easy identification, therefore, the character picture after the present invention adopts simply and effectively feature is cut each carries out " U " identification, and concrete steps are as follows:
Referring to Fig. 6, at the character picture of well cutting bottom three/part scanning straight line, if picture below straight line exists, then preliminary judgement is " U ", otherwise thinks character non-" U ".
At preliminary judgement is on the character picture of " U ", flash trimming out-of-bounds three minutes highly on the statistics saltus step count, count in the statistics saltus step of character picture width equinox place, satisfy 4,4,2 if three saltus steps are counted, think that then character " U " exists.
(5) case number (CN) character recognition and verification
Extract character feature one by one, adopt algorithm simple on the feature classification, the accurate method that combines with architectural feature of the statistical nature of identification character is again verified identification case number (CN) in conjunction with the verification criterion simultaneously, guarantee the high efficiency and the accuracy of case number (CN) identification, its process is as follows:
Extract 0 ~ 90 arabic numeral from cut apart good container code characters, totally 26 English alphabets of A ~ Z carries out size normalization with each character, and place normalization becomes the 42*28 pixel, does refinement again and handles, and makes template.
Each character is divided into six grids equably, add up character pixels number in each grid, form sextuple proper vector, draw bar line computation point of crossing number in the vertical orientation of character again, horizontal trisection posistion is drawn two horizontal lines and is calculated the point of crossing number, forms nine dimensional feature vectors so altogether.
36 characters in the template are carried out as above feature extraction, form the feature database of 36*9.
Extract the feature of character to be identified according to above-mentioned feature extracting method, English alphabet is separated with numeral discern, its identification principle is according to the method for Weighted distance template matches, and formula is as follows:
Feature and template base Chinese and English letter (numeral) feature of English alphabet to be identified (numeral) are calculated Weighted distance according to above-mentioned formula, minimum with the distance of which template character, then just be this template character with character recognition to be identified.Experiment shows that a gets 2, and b gets 3, and accuracy is the highest.
After identifying every character, use the case number (CN) check bit that identification case number (CN) is tested.The case number (CN) check bit i.e. the 11st position case number (CN), and it is to be obtained through certain verification rule computing by preceding 4 English alphabets and back 6 bit digital.Behind 11 case number (CN)s of identification, adopt certain mathematical operation that the pairing numeral of top ten case number (CN) is carried out computing by case number (CN) verification criterion, the gained result is compared with the check bit that identifies, if consistent, then case number (CN) is discerned successfully; Otherwise jump to character " U " identifying.
Based on the container number automatic identifying method that such scheme forms, it specifically implements as follows:
Referring to Fig. 2, this example relates to case number (CN) that a width of cloth obtains by the camera case number (CN) picture processing procedure for " CAXU 741,193 8 ":
Image inclination is proofreaied and correct
Referring to Fig. 3, after obtaining bitmap, grasp casing frame straight line by the Hough conversion, usually choose casing coboundary and right margin (comparing) with Fig. 2, and be that reference line is done slant correction with this frame, be that benchmark extends into prescribed level in the below left with two borders again, intercept it, to do further processing.
The image pre-service
Obtain and proofread and correct the back picture, obtain the gradation of image value by the gradation conversion formula:
Wherein Grey is the gradation of image value after changing, and R, G, B are respectively redness, green and the blue component of original color image.
By dynamic threshold formula is set:
Wherein
Be grey scale pixel value,
Be respectively and get average and mean square deviation computing.
Have some chamber door vertical metal bars for the container number picture that obtains from the case tail and disturb, these are very disadvantageous for the case number (CN) leaching process, therefore by row to the scan statistics mark tally, if then thinking to exist greater than threshold value disturbs, in addition filtering.
In the case number (CN) picture that case tail and case side photograph, its raising middle flask tail picture disturbs at most, this not only comprises the interference of external object, the character of container type information also on the books simultaneously disturbs, little more a lot of on these character individualities than 11 case number (CN)s, so adopt morphological image algorithm opening operation in this example:
Wherein S represents to carry out the bianry image set of opening operation, and B represents to be used for carrying out the structural element of opening operation, and X representation element image is through the collection of pixels after the binaryzation.
Represent that respectively opening operation, burn into expand.
Come these small characters of filtering, this operation also helps strengthening the treatment effect of follow up scan separating character.
Referring to Fig. 4, it is case number (CN) picture pretreating effect figure, and from scheming to go up visible chamber door Metallic rod by filtering, wherein individual less its signal to noise ratio (S/N ratio) of interference character also reduces gradually.
The rank scanning method is cut apart " case number (CN) character "
Only have a few regions to keep higher signal noise ratio (snr) of image through above-mentioned pretreated case number (CN) picture, this is comprising 11 case number (CN) characters and 4 heavy information characters of casees, as " 42G1 " etc.By the thought of horizontal projection method, will orient the character zone that satisfies the " char " condition, each " char " splits in will going by vertical projection method's thought simultaneously, will obtain character: 42G1CAXU7411938(as shown in Figure 5).
EIC equipment identification code " U " is judged
The character that splits generally speaking can comprise some box information characters, and as " 42G1 " and class character icon, these characters disturb the extraction identifying of actual case number (CN) character, therefore needs effectively to remove non-case number (CN) and disturbs.
And in case number (CN) was formed, the most tangible characteristics were EIC equipment identification code " U ".
Referring to Fig. 6, this example is by the statistics intersection feature, and by every width of cloth picture of intercepting is added up simple intersection feature positioning equipment identification code " U ", this will improve whole system operation efficient effectively.
Character recognition
For guaranteeing compatibility, the portability of algorithm and hardware, present embodiment adopts the relatively low masterplate matching algorithm of computation complexity to realize the identifying of case number (CN) character.
At the identification preprocessing part, adopt the dilation operation in the morphology that the fracture character is repaired.
In characteristic aspect, the statistical nature (grid feature, projection properties etc.) of extraction character and architectural feature (end points, triradius, four crunode numbers, closed loop number) etc., by carrying out Distance Matching with the template characteristic storehouse, finish identifying, identify case number (CN) character: CAXU 741193 8(the most at last as shown in Figure 7).
Thus, recognition methods provided by the invention can identify container number fast, accurately.After identifying finishes, this algorithm will be verified identification case number (CN) in conjunction with case number (CN) verification rule, guarantee identification accuracy.
Simultaneously, the present invention is all effective for the case number (CN) character of various arrangements such as delegation, two row, four lines, row, therefore can to casing each to carrying out case number (CN) identification, add up everywhere that the result is verified, improve the accuracy of case number (CN) location, identification.
More than show and described ultimate principle of the present invention, principal character and advantage of the present invention.The technician of the industry should understand; the present invention is not restricted to the described embodiments; that describes in the foregoing description and the instructions just illustrates principle of the present invention; without departing from the spirit and scope of the present invention; the present invention also has various changes and modifications, and these changes and improvements all fall in the claimed scope of the invention.The claimed scope of the present invention is defined by appending claims and equivalent thereof
Claims (5)
1. the container number automatic identifying method is characterized in that, described automatic identifying method may further comprise the steps:
(1) image inclination is proofreaied and correct: after obtaining the case number (CN) picture, search set vanning frame is thought line of reference measuring and calculating degree of tilt with this frame, proofreaies and correct picture in the back intercepting frame;
(2) image pre-service: with truncated picture gray processing, dynamically binaryzation, filtering noise interference;
(3) dividing box sign character:, scan and be partitioned into " case number (CN) character " picture that satisfies condition one by one by the rank scanning method;
(4) determining apparatus identification code:, accurately extract 11 case number (CN) characters that satisfy condition by the determining apparatus identification code;
(5) case number (CN) character recognition and verification: extract character feature one by one, finish identifying, judge in conjunction with the verification criterion whether case number (CN) identification is successful.
2. container number automatic identifying method according to claim 1 is characterized in that, the image pre-service of described step (2) realizes by following steps:
(2.1) gray processing
Read the progress R of the pixel in the data bitmap district of the image of (1) processing suddenly, G, the B tristimulus values is converted into gray-scale map by the gradation conversion formula with picture:
Wherein Grey is the gradation of image value after changing, and R, G, B are respectively redness, green and the blue component of original color image;
(2.2) dynamic binaryzation
Add up the bitmap pixels gray average and the mean square deviation of gray processing, calculate the dynamic binary-state threshold of view picture bitmap by following formula:
Wherein
Be grey scale pixel value,
Be respectively and get average and mean square deviation computing;
After obtaining this threshold value, image is carried out the binaryzation process, if grey scale pixel value greater than this threshold value, then is set at 255 with this grey scale pixel value, otherwise is 0;
(3.3) filtering noise
Utilize door lock Metallic rod interference in the linear feature filtering image: after the binaryzation, by the Hough conversion, the thread of scan image, when the row of straight line surpass certain threshold value to the foreground point pixel count, think that this straight line is that Metallic rod is listed as to the border, when the spacing of two metal line during less than certain threshold value, assert that the line-to-line zone is the Metallic rod district, this area pixel gray-scale value is made as 0;
Adopt morphological image algorithm opening operation to desalinate such character gradually, reduce to disturb: its formula is as follows:
Wherein S represents to carry out the bianry image set of opening operation, and B represents to be used for carrying out the structural element of opening operation, the collection of pixels after the X representation element image process binaryzation,
Represent that respectively opening operation, burn into expand.
3. container number automatic identifying method according to claim 1 is characterized in that, the dividing box sign character of described step (3) is realized by following steps:
(3.1) by the case number (CN) picture of lining by line scan, the searching image prospect is counted, and when surpassing certain threshold value when counting, tentatively regards as character zone, and makes mark, and whether need to distinguish one's own profession in this process is the up-and-down boundary of character zone;
(3.2) scanning entire image is demarcated all row zones that comprises character;
(3.3) after the line scanning, to demarcating the row zone picture is listed as to scanning, when being listed as when image is counted above certain threshold value, character exists, and by edge determination, demarcates its border, left and right sides, simultaneously in conjunction with going regional this character of cutting.
4. container number automatic identifying method according to claim 1 is characterized in that, the determining apparatus identification code of described step (4) realizes by following steps:
(4.1) at the character picture of well cutting bottom three/part scanning straight line, if picture below straight line exists, then preliminary judgement is EIC equipment identification code " U ", otherwise thinks the non-EIC equipment identification code of character " U ";
(4.2) be on the character picture of " U " at preliminary judgement, flash trimming out-of-bounds three minutes highly on the statistics saltus step count, count in the statistics saltus step of character picture width equinox place, satisfy 4,4,2 if three saltus steps are counted, think that then character " U " exists.
5. container number automatic identifying method according to claim 1 and verification decision process is characterized in that, the case number (CN) character recognition of described step (5) realizes by following steps:
(5.1) extract 0 ~ 90 arabic numeral from cut apart good container code characters, totally 26 English alphabets of A ~ Z carries out size normalization with each character, and place normalization becomes the 42*28 pixel, does refinement again and handles, and makes template;
(5.2) each character is divided into equably six grids, add up character pixels number in each grid, form sextuple proper vector, draw bar line computation point of crossing number in the vertical orientation of character again, horizontal trisection posistion is drawn two horizontal lines and is calculated the point of crossing number, forms nine dimensional feature vectors altogether;
(5.3) 36 characters in the template are carried out feature extraction according to step (5.2), form the feature database of 36*9;
(5.4) extract the feature of character to be identified according to above-mentioned feature extracting method, English alphabet is separated with numeral discern, its identification principle is according to the method for Weighted distance template matches, and formula is as follows:
The feature of English alphabet to be identified and numeral is calculated Weighted distance with template base Chinese and English letter and number feature according to above-mentioned formula respectively, minimum with the distance of which template character, then just be this template character with character recognition to be identified;
(5.5) after identifying every character, use the case number (CN) check bit that identification case number (CN) is tested.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010554679 CN102024144A (en) | 2010-11-23 | 2010-11-23 | Container number identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010554679 CN102024144A (en) | 2010-11-23 | 2010-11-23 | Container number identification method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102024144A true CN102024144A (en) | 2011-04-20 |
Family
ID=43865425
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010554679 Pending CN102024144A (en) | 2010-11-23 | 2010-11-23 | Container number identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102024144A (en) |
Cited By (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102184390A (en) * | 2011-05-17 | 2011-09-14 | 姜雨枫 | Container number-orientated character image identification method |
| CN102306283A (en) * | 2011-07-23 | 2012-01-04 | 浙江大学 | Image identification method of indicating values of LCD (liquid crystal display) digital display instrument |
| CN104376322A (en) * | 2014-12-01 | 2015-02-25 | 上海海事大学 | Intelligent detecting and evaluating method for container number preprocessing quality of containers |
| CN104601877A (en) * | 2014-05-08 | 2015-05-06 | 腾讯科技（深圳）有限公司 | Image processing method, device and system |
| CN104820827A (en) * | 2015-04-28 | 2015-08-05 | 电子科技大学 | Method for recognizing punctiform characters on surfaces of cables |
| CN105512607A (en) * | 2015-11-25 | 2016-04-20 | 浙江工业大学 | Identification method for container box truck lock head positions |
| CN105701490A (en) * | 2016-02-24 | 2016-06-22 | 上海海事大学 | Container number adaptive positioning method based on image entropy |
| CN105761351A (en) * | 2016-01-08 | 2016-07-13 | 东方通信股份有限公司 | Structure characteristic-based character recognition method |
| CN105787894A (en) * | 2016-02-25 | 2016-07-20 | 上海海事大学 | Barrel distortion container number correction method |
| CN105844205A (en) * | 2015-01-15 | 2016-08-10 | 新天科技股份有限公司 | Character information identification method based on image processing |
| CN105913059A (en) * | 2016-04-11 | 2016-08-31 | 吉林大学 | Vehicle VIN code automatic identifying system and control method therefor |
| CN106203239A (en) * | 2015-05-04 | 2016-12-07 | 杭州海康威视数字技术股份有限公司 | Information processing method, device and system for container tally |
| CN106203539A (en) * | 2015-05-04 | 2016-12-07 | 杭州海康威视数字技术股份有限公司 | The method and apparatus identifying container number |
| CN106803087A (en) * | 2017-01-19 | 2017-06-06 | 南京航空航天大学 | A kind of car number automatic identification method and system |
| CN107093059A (en) * | 2017-03-31 | 2017-08-25 | 南京钧元网络科技有限公司 | A kind of portable authentication intelligent checks system and its checking method |
| CN107832767A (en) * | 2017-11-14 | 2018-03-23 | 深圳码隆科技有限公司 | Container number identification method, device and electronic equipment |
| CN108036778A (en) * | 2017-12-26 | 2018-05-15 | 清华大学 | Container automatic recognition system and recognition methods and port facilities |
| CN108038482A (en) * | 2017-12-22 | 2018-05-15 | 大连运明自动化技术有限公司 | A kind of automobile engine cylinder-body sequence number Visual intelligent identifying system |
| CN108304836A (en) * | 2018-02-01 | 2018-07-20 | 上海海事大学 | A kind of books Ordering Detection Algorithm based on image |
| CN108470366A (en) * | 2018-03-28 | 2018-08-31 | 同方威视技术股份有限公司 | Analog image generation method and device and computer readable storage medium |
| CN108805129A (en) * | 2017-08-24 | 2018-11-13 | 贵州省烟草公司贵阳市公司 | Tobacco laser code assists in identifying method, apparatus and tobacco laser code identification equipment |
| CN109165649A (en) * | 2018-09-03 | 2019-01-08 | 苏州巨能图像检测技术有限公司 | The high accurate detection method in the container case hole of view-based access control model detection |
| CN109190625A (en) * | 2018-07-06 | 2019-01-11 | 同济大学 | A kind of container number identification method of wide-angle perspective distortion |
| CN110119742A (en) * | 2019-04-25 | 2019-08-13 | 添维信息科技(天津)有限公司 | A kind of recognition methods of container number, device and mobile terminal |
| CN110321884A (en) * | 2019-06-13 | 2019-10-11 | 贝式计算(天津)信息技术有限公司 | Number the method and device of identification |
| CN110348451A (en) * | 2019-07-18 | 2019-10-18 | 西南交通大学 | Case number (CN) automatic collection and recognition methods in railway container cargo handling process |
| CN110378332A (en) * | 2019-06-14 | 2019-10-25 | 上海咪啰信息科技有限公司 | A kind of container terminal case number (CN) and Train number recognition method and system |
| CN110648540A (en) * | 2019-09-09 | 2020-01-03 | 安徽师范大学 | Expressway emergency lane occupation tracking system and method based on unmanned aerial vehicle |
| CN111027546A (en) * | 2019-12-05 | 2020-04-17 | 北京嘉楠捷思信息技术有限公司 | Character segmentation method and device and computer readable storage medium |
| CN111079735A (en) * | 2018-10-22 | 2020-04-28 | 莱芜钢铁集团电子有限公司 | Method and device for recognizing end face characters of casting blank |
| CN111243016A (en) * | 2018-11-28 | 2020-06-05 | 周口师范学院 | Automatic identification and positioning method for container |
| CN113971811A (en) * | 2021-11-16 | 2022-01-25 | 北京国泰星云科技有限公司 | Intelligent container feature identification method based on machine vision and deep learning |
| CN114821134A (en) * | 2022-06-30 | 2022-07-29 | 山东蓝彩天下教育科技有限公司 | Method for identifying print style number of publication based on template matching |
| CN116452467A (en) * | 2023-06-16 | 2023-07-18 | 山东曙岳车辆有限公司 | Container real-time positioning method based on laser data |
| CN116912845A (en) * | 2023-06-16 | 2023-10-20 | 广东电网有限责任公司佛山供电局 | Intelligent content identification and analysis method and device based on NLP and AI |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI231918B (en) * | 2003-03-27 | 2005-05-01 | Chunghwa Telecom Co Ltd | Automatic container number recognition method enabling color separation and result collection |
| US7116814B2 (en) * | 2002-09-27 | 2006-10-03 | Chunghwa Telecom Co., Ltd. | Image-based container defects detector |
| CN101650780A (en) * | 2009-09-08 | 2010-02-17 | 宁波中科集成电路设计中心有限公司 | Identification method of container number |
-
2010
- 2010-11-23 CN CN 201010554679 patent/CN102024144A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7116814B2 (en) * | 2002-09-27 | 2006-10-03 | Chunghwa Telecom Co., Ltd. | Image-based container defects detector |
| TWI231918B (en) * | 2003-03-27 | 2005-05-01 | Chunghwa Telecom Co Ltd | Automatic container number recognition method enabling color separation and result collection |
| CN101650780A (en) * | 2009-09-08 | 2010-02-17 | 宁波中科集成电路设计中心有限公司 | Identification method of container number |
Non-Patent Citations (1)
| Title |
|---|
| 《中国优秀硕士学位论文全文数据库》 20040202 卢盛辉 集装箱自动识别***研究 , * |
Cited By (57)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102184390B (en) * | 2011-05-17 | 2013-01-23 | 姜雨枫 | Container number-orientated character image identification method |
| CN102184390A (en) * | 2011-05-17 | 2011-09-14 | 姜雨枫 | Container number-orientated character image identification method |
| CN102306283A (en) * | 2011-07-23 | 2012-01-04 | 浙江大学 | Image identification method of indicating values of LCD (liquid crystal display) digital display instrument |
| CN102306283B (en) * | 2011-07-23 | 2013-06-12 | 浙江大学 | Image identification method of indicating values of LCD (liquid crystal display) digital display instrument |
| CN104601877B (en) * | 2014-05-08 | 2017-10-10 | 腾讯科技（深圳）有限公司 | Image processing method, apparatus and system |
| CN104601877A (en) * | 2014-05-08 | 2015-05-06 | 腾讯科技（深圳）有限公司 | Image processing method, device and system |
| CN104376322A (en) * | 2014-12-01 | 2015-02-25 | 上海海事大学 | Intelligent detecting and evaluating method for container number preprocessing quality of containers |
| CN104376322B (en) * | 2014-12-01 | 2017-11-17 | 上海海事大学 | A kind of intelligent evaluating method of container number pretreatment quality |
| CN105844205B (en) * | 2015-01-15 | 2019-05-31 | 新天科技股份有限公司 | Character information recognition methods based on image procossing |
| CN105844205A (en) * | 2015-01-15 | 2016-08-10 | 新天科技股份有限公司 | Character information identification method based on image processing |
| CN104820827A (en) * | 2015-04-28 | 2015-08-05 | 电子科技大学 | Method for recognizing punctiform characters on surfaces of cables |
| CN106203539B (en) * | 2015-05-04 | 2020-01-17 | 杭州海康威视数字技术股份有限公司 | Method and device for identifying container number |
| CN106203239B (en) * | 2015-05-04 | 2020-08-04 | 杭州海康威视数字技术股份有限公司 | Information processing method, device and system for container tallying |
| CN106203239A (en) * | 2015-05-04 | 2016-12-07 | 杭州海康威视数字技术股份有限公司 | Information processing method, device and system for container tally |
| CN106203539A (en) * | 2015-05-04 | 2016-12-07 | 杭州海康威视数字技术股份有限公司 | The method and apparatus identifying container number |
| CN105512607B (en) * | 2015-11-25 | 2019-01-15 | 浙江工业大学 | A kind of truck tapered end location recognition method |
| CN105512607A (en) * | 2015-11-25 | 2016-04-20 | 浙江工业大学 | Identification method for container box truck lock head positions |
| CN105761351A (en) * | 2016-01-08 | 2016-07-13 | 东方通信股份有限公司 | Structure characteristic-based character recognition method |
| CN105701490B (en) * | 2016-02-24 | 2018-11-30 | 上海海事大学 | A kind of container number adaptive location method based on image entropy |
| CN105701490A (en) * | 2016-02-24 | 2016-06-22 | 上海海事大学 | Container number adaptive positioning method based on image entropy |
| CN105787894A (en) * | 2016-02-25 | 2016-07-20 | 上海海事大学 | Barrel distortion container number correction method |
| CN105913059A (en) * | 2016-04-11 | 2016-08-31 | 吉林大学 | Vehicle VIN code automatic identifying system and control method therefor |
| CN105913059B (en) * | 2016-04-11 | 2021-04-27 | 吉林大学 | Automatic identification system for vehicle VIN code and control method thereof |
| CN106803087A (en) * | 2017-01-19 | 2017-06-06 | 南京航空航天大学 | A kind of car number automatic identification method and system |
| CN107093059A (en) * | 2017-03-31 | 2017-08-25 | 南京钧元网络科技有限公司 | A kind of portable authentication intelligent checks system and its checking method |
| CN108805129A (en) * | 2017-08-24 | 2018-11-13 | 贵州省烟草公司贵阳市公司 | Tobacco laser code assists in identifying method, apparatus and tobacco laser code identification equipment |
| CN107832767A (en) * | 2017-11-14 | 2018-03-23 | 深圳码隆科技有限公司 | Container number identification method, device and electronic equipment |
| CN108038482A (en) * | 2017-12-22 | 2018-05-15 | 大连运明自动化技术有限公司 | A kind of automobile engine cylinder-body sequence number Visual intelligent identifying system |
| CN108036778B (en) * | 2017-12-26 | 2024-04-02 | 清华大学 | Container automatic identification system and identification method and port facility |
| CN108036778A (en) * | 2017-12-26 | 2018-05-15 | 清华大学 | Container automatic recognition system and recognition methods and port facilities |
| CN108304836A (en) * | 2018-02-01 | 2018-07-20 | 上海海事大学 | A kind of books Ordering Detection Algorithm based on image |
| CN108470366A (en) * | 2018-03-28 | 2018-08-31 | 同方威视技术股份有限公司 | Analog image generation method and device and computer readable storage medium |
| CN109190625A (en) * | 2018-07-06 | 2019-01-11 | 同济大学 | A kind of container number identification method of wide-angle perspective distortion |
| CN109190625B (en) * | 2018-07-06 | 2021-09-03 | 同济大学 | Large-angle perspective deformation container number identification method |
| CN109165649B (en) * | 2018-09-03 | 2022-04-15 | 苏州巨能图像检测技术有限公司 | High-precision container hole detection method based on visual detection |
| CN109165649A (en) * | 2018-09-03 | 2019-01-08 | 苏州巨能图像检测技术有限公司 | The high accurate detection method in the container case hole of view-based access control model detection |
| CN111079735B (en) * | 2018-10-22 | 2023-12-22 | 莱芜钢铁集团电子有限公司 | Casting blank end face character recognition method and device |
| CN111079735A (en) * | 2018-10-22 | 2020-04-28 | 莱芜钢铁集团电子有限公司 | Method and device for recognizing end face characters of casting blank |
| CN111243016B (en) * | 2018-11-28 | 2024-03-19 | 周口师范学院 | Automatic container identification and positioning method |
| CN111243016A (en) * | 2018-11-28 | 2020-06-05 | 周口师范学院 | Automatic identification and positioning method for container |
| CN110119742B (en) * | 2019-04-25 | 2023-07-07 | 添维信息科技(天津)有限公司 | Container number identification method and device and mobile terminal |
| CN110119742A (en) * | 2019-04-25 | 2019-08-13 | 添维信息科技(天津)有限公司 | A kind of recognition methods of container number, device and mobile terminal |
| CN110321884B (en) * | 2019-06-13 | 2020-11-13 | 贝式计算(天津)信息技术有限公司 | Method and device for identifying serial number |
| CN110321884A (en) * | 2019-06-13 | 2019-10-11 | 贝式计算(天津)信息技术有限公司 | Number the method and device of identification |
| CN110378332A (en) * | 2019-06-14 | 2019-10-25 | 上海咪啰信息科技有限公司 | A kind of container terminal case number (CN) and Train number recognition method and system |
| CN110348451A (en) * | 2019-07-18 | 2019-10-18 | 西南交通大学 | Case number (CN) automatic collection and recognition methods in railway container cargo handling process |
| CN110648540B (en) * | 2019-09-09 | 2021-07-02 | 安徽师范大学 | Expressway emergency lane occupation tracking system and method based on unmanned aerial vehicle |
| CN110648540A (en) * | 2019-09-09 | 2020-01-03 | 安徽师范大学 | Expressway emergency lane occupation tracking system and method based on unmanned aerial vehicle |
| CN111027546A (en) * | 2019-12-05 | 2020-04-17 | 北京嘉楠捷思信息技术有限公司 | Character segmentation method and device and computer readable storage medium |
| CN111027546B (en) * | 2019-12-05 | 2024-03-26 | 嘉楠明芯(北京)科技有限公司 | Character segmentation method, device and computer readable storage medium |
| CN113971811A (en) * | 2021-11-16 | 2022-01-25 | 北京国泰星云科技有限公司 | Intelligent container feature identification method based on machine vision and deep learning |
| CN114821134B (en) * | 2022-06-30 | 2022-09-02 | 山东蓝彩天下教育科技有限公司 | Method for identifying print style number of publication based on template matching |
| CN114821134A (en) * | 2022-06-30 | 2022-07-29 | 山东蓝彩天下教育科技有限公司 | Method for identifying print style number of publication based on template matching |
| CN116452467A (en) * | 2023-06-16 | 2023-07-18 | 山东曙岳车辆有限公司 | Container real-time positioning method based on laser data |
| CN116452467B (en) * | 2023-06-16 | 2023-09-22 | 山东曙岳车辆有限公司 | Container real-time positioning method based on laser data |
| CN116912845A (en) * | 2023-06-16 | 2023-10-20 | 广东电网有限责任公司佛山供电局 | Intelligent content identification and analysis method and device based on NLP and AI |
| CN116912845B (en) * | 2023-06-16 | 2024-03-19 | 广东电网有限责任公司佛山供电局 | Intelligent content identification and analysis method and device based on NLP and AI |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102024144A (en) | Container number identification method | |
| Babu et al. | Vehicle number plate detection and recognition using bounding box method | |
| CN108596166A (en) | A kind of container number identification method based on convolutional neural networks classification | |
| Ashtari et al. | An Iranian license plate recognition system based on color features | |
| CN105373794B (en) | A kind of licence plate recognition method | |
| CN112686812B (en) | Bank card inclination correction detection method and device, readable storage medium and terminal | |
| US10430681B2 (en) | Character segmentation and recognition method | |
| Prabhakar et al. | Automatic vehicle number plate detection and recognition | |
| CN102163284B (en) | Chinese environment-oriented complex scene text positioning method | |
| CN103034848B (en) | A kind of recognition methods of form types | |
| CN103824091B (en) | A kind of licence plate recognition method for intelligent transportation system | |
| CN106875546A (en) | A kind of recognition methods of VAT invoice | |
| CN106529532A (en) | License plate identification system based on integral feature channels and gray projection | |
| CN108171127A (en) | A kind of invoice automatic identifying method based on deep learning | |
| CN104700092A (en) | Small-character number identification method based on template and feature matching | |
| CN106503638B (en) | Image procossing, vehicle color identification method and system for color identification | |
| Prabhakar et al. | A novel design for vehicle license plate detection and recognition | |
| CN108921152A (en) | English character cutting method and device based on object detection network | |
| Shivakumara et al. | Gradient-angular-features for word-wise video script identification | |
| Chandra et al. | Recognition of vehicle number plate using Matlab | |
| CN109460722A (en) | A kind of license plate intelligent identification Method | |
| Suresh et al. | Indian currency recognition and verification using image processing | |
| Choudhury et al. | A new zone based algorithm for detection of license plate from Indian vehicle | |
| Dhar et al. | Bangladeshi license plate recognition using adaboost classifier | |
| CN107392115B (en) | Traffic sign identification method based on hierarchical feature extraction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110420 |