CN117115701A - Platform truck operation behavior identification method for customs auxiliary management - Google Patents
Platform truck operation behavior identification method for customs auxiliary management Download PDFInfo
- Publication number
- CN117115701A CN117115701A CN202310906974.5A CN202310906974A CN117115701A CN 117115701 A CN117115701 A CN 117115701A CN 202310906974 A CN202310906974 A CN 202310906974A CN 117115701 A CN117115701 A CN 117115701A
- Authority
- CN
- China
- Prior art keywords
- image
- license plate
- platform
- truck
- column
- 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
- 238000000034 method Methods 0.000 title claims abstract description 31
- 239000011159 matrix material Substances 0.000 claims description 38
- 230000009467 reduction Effects 0.000 claims description 13
- 238000003708 edge detection Methods 0.000 claims description 9
- 238000001514 detection method Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 230000003321 amplification Effects 0.000 claims description 3
- 230000008602 contraction Effects 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 3
- 238000004891 communication Methods 0.000 claims description 2
- 230000006399 behavior Effects 0.000 claims 9
- 241001270131 Agaricus moelleri Species 0.000 claims 1
- 230000005540 biological transmission Effects 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 230000007306 turnover Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/457—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by analysing connectivity, e.g. edge linking, connected component analysis or slices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/54—Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
-
- 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
- G06V20/625—License plates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/15—Cutting or merging image elements, e.g. region growing, watershed or clustering-based techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a platform truck operation behavior identification method for customs auxiliary management, which belongs to the technical field of license plate identification and specifically comprises the following steps: establishing a cloud platform, wherein the cloud platform is used for storing truck registration information and a platform scheduling plan; acquiring a video image of a platform parking space, acquiring an occupied state of the platform parking space according to the video image, and uploading the occupied state to the cloud platform; identifying an entrance truck, judging whether the truck is a registered truck, and distributing a parking space for the truck according to the platform dispatching plan and the occupied state; identifying license plate information of a truck occupying a parking space through the video image, judging whether the truck accords with the platform scheduling information, if so, carrying out platform operation, and if not, carrying out prompt; the application realizes accurate identification of license plate information.
Description
Technical Field
The application relates to the technical field of license plate recognition, in particular to a platform truck operation behavior recognition method for customs auxiliary management.
Background
Currently, in order to increase the turnover efficiency of a logistics park, trucks in the park are usually parked on a distribution platform. And the occupancy state of the dock is determined through the dock camera, and the real-time dispatching of the trucks in the logistics park is realized by means of the identification of the dock camera to the license plates of the trucks occupying the dock, so as to judge whether the trucks are parked at the dock allocated for the logistics park. If the truck is determined to stop at the wrong dock, a dispatcher is required to contact a driver to drive the truck away from the dock;
in a real license plate recognition system, license plate images are often polluted by various noises due to imperfections of an imaging system, a transmission medium, recording equipment and the like, changes of weather conditions and the like, so that the difficulty of accurately recognizing the license plate is high.
Disclosure of Invention
The application aims to provide a platform truck operation behavior identification method for customs auxiliary management, which solves the following technical problems:
(1) In a real license plate recognition system, license plate images are often polluted by various noises due to imperfections of an imaging system, a transmission medium, recording equipment and the like, changes of weather conditions and the like, so that the difficulty of accurately recognizing the license plate is high.
The aim of the application can be achieved by the following technical scheme:
a dock wagon operation behavior identification method for customs assistance management, comprising:
establishing a cloud platform, wherein the cloud platform is used for storing truck registration information and a platform scheduling plan;
acquiring a video image of a platform parking space, acquiring an occupied state of the platform parking space according to the video image, and uploading the occupied state to the cloud platform;
identifying an entrance truck, judging whether the truck is a registered truck, and distributing a parking space for the truck according to the platform dispatching plan and the occupied state;
and identifying license plate information of the truck occupying the parking space through the video image, judging whether the truck accords with the platform scheduling information, if so, carrying out platform operation, and if not, carrying out prompt.
As a further scheme of the application: the process for identifying the license plate information comprises the following steps:
splitting the video image into a plurality of image frames, carrying out gray processing on the image frames to obtain gray images, and carrying out noise reduction processing on the gray images.
As a further scheme of the application: and carrying out edge detection on the gray level image after noise reduction by adopting a Sobel edge detection operator, carrying out binarization processing on the gray level image after the edge detection by adopting a maximum inter-class variance method to obtain a binarized image, wherein the pixel value of the gray level image after the binarization processing is 0 or 255, and carrying out noise reduction processing on the binarized image again.
As a further scheme of the application: selecting the outline of a white communication area in a binary image through a rectangular boundary frame to obtain a rectangular area to be determined, wherein the aspect ratio of the rectangular boundary frame is A+/-a:B+/-B, A, B is a preset coefficient, a and B are preset error thresholds, the rectangular area to be determined is subjected to expansion and amplification, if a plurality of rectangular areas to be determined exist, the pixel ratio of gray values of 255 and 0 in the rectangular area to be determined and the jump frequency range of gray in the rectangular area to be determined are respectively calculated, if the pixel ratio of the rectangular area to be determined is lower than 0.25 and the jump frequency range is [5,30], the rectangular area to be determined to be a license plate image, otherwise, the selection is continued.
As a further scheme of the application: the license plate image is segmented by license plate characters, two thresholds T1 and T2 are set, the license plate image is scanned from left to right, a first column with the pixel value of 255 being larger than the threshold T1 is obtained, namely a starting column of license plate Chinese characters is marked as S, the license plate image is continuously scanned until the column with the pixel value of 255 being smaller than the threshold T1 is identified, O is marked as O, if O-S is larger than T2, O is the ending column of license plate Chinese characters, otherwise, the image is continuously scanned until the column with the width being larger than T2 from S is identified, and the number of pixels with the pixel value of 255 is smaller than the preset threshold, and the column is the ending column of license plate Chinese characters;
and continuing scanning the rest characters, obtaining a column with the pixel value of 255 and the pixel number of more than the threshold value T1 as a character starting column, and repeating the steps by taking a column with the pixel value of 255 and the pixel number of less than the threshold value T1 as a character ending column to finish character segmentation of the license plate.
As a further scheme of the application: the cloud platform also comprises a character library, wherein the character library is generated based on big data, standard characters of various license plates are stored in the character library, the standard characters are selected, the outer edge lines of the standard characters are drawn, a coordinate system is established by taking the geometric center point of the standard characters as an origin, a plurality of pairs of characteristic points on the outer edge lines of the standard characters are selected as reference points, the reference points are marked as (Xi, yi) and (Mi, ni), and i= … n;
selecting the license plate character, drawing the outer edge line of the license plate character, establishing a coordinate system by taking the geometric center point of the character as an origin, and selecting a plurality of pairs of characteristic points on the outer edge line of the license plate character as detection points, wherein the detection points are marked as (Ui, vi) and (Zi, wi), and i= … n;
by the formulaObtaining a similarity value XS of license plate characters and standard characters, judging the license plate characters as corresponding standard characters if the similarity value XS is lower than a preset threshold value, obtaining license plate information, and otherwise, continuing to identify.
As a further scheme of the application: the noise reduction process comprises the following steps:
dividing the gray image into a plurality of overlapped image blocks, taking the image blocks around each image block, calculating and storing norms among the image blocks, and then carrying out non-dominant sorting, wherein the norms comprise MSE and MAE.
As a further scheme of the application: the specific calculation process of the norm comprises the following steps:
selecting image blocks and marking as P 1 Obtaining P 1 And marked as P 2 Block P of picture 1 Conversion to a formal representation of matrix i, converting image block P 2 Is converted into a form of matrix j, corresponding elements of the matrix i and the matrix j are subtracted, the sum of squares of the differences is calculated, and the sum is divided by the size of the corresponding matrix, thus obtaining an image block P 1 And image block P 2 Subtracting the corresponding elements of matrix i and matrix j and calculating the sum of the absolute values of the differences, dividing the sum by the size of the matrix to obtain the image block P 1 And adjacent image block P 2 Is a MAE of (c).
As a further scheme of the application: the top k image blocks P in the non-dominant order 2 Forming a matrix, and performing soft threshold contraction on singular values of the matrix by using different lambda values to obtain a low-rank matrix;
and acquiring singular values of the low-rank matrix, filtering smaller numerical values in the singular values, and merging the obtained noise-reduced image blocks into a noise-reduced gray image.
The application has the beneficial effects that:
(1) In order to quickly and accurately position the license plate under the complex background and uneven illumination conditions, the application solves the problems of the license plate positioning algorithm based on an improved Sobel edge detection operator, realizes the extraction of texture features of the license plate image in horizontal, vertical and diagonal directions, binarizes the license plate image by adopting a maximum inter-class variance method algorithm, carries out mathematical morphological operation on the binary image to obtain a candidate region of the license plate, removes a fake license plate by utilizing the license plate features, segments license plate characters by a threshold method and identifies the characters by feature comparison;
(2) According to the application, non-dominant ordering is utilized to replace the existing Euclidean distance method, similarity between image blocks is calculated through two norms, the non-dominant ordering of the two norms is carried out on the image blocks, the characteristic that a similar block matrix of a noise-free image is embodied as a low-rank matrix is adopted, isolated edge pixel points are restrained through a Sobel operator, the image blocks are matched with high efficiency, and license plate information is conveniently identified.
Drawings
The application is further described below with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Referring to fig. 1, the application provides a platform wagon operation behavior identification method for customs auxiliary management, comprising the following steps:
establishing a cloud platform, wherein the cloud platform is used for storing truck registration information and a platform scheduling plan;
acquiring a video image of a platform parking space, acquiring an occupied state of the platform parking space according to the video image, and uploading the occupied state to the cloud platform;
identifying an entrance truck, judging whether the truck is a registered truck, and distributing a parking space for the truck according to the platform dispatching plan and the occupied state;
and identifying license plate information of the truck occupying the parking space through the video image, judging whether the truck accords with the platform scheduling information, if so, carrying out platform operation, and if not, carrying out prompt.
In a real license plate recognition system, because of imperfections of an imaging system, a transmission medium, recording equipment and the like, changes of weather conditions and the like, license plate images are often polluted by various noises, and the difficulty of accurately positioning the license plate is high, an image noise reduction method effectively applied to the field of platform truck recognition is needed; judging whether a parking space at a platform is in an occupied state or not according to video image data information, if so, identifying license plate information of a truck corresponding to the occupied parking space, and matching the license plate information of the truck corresponding to the occupied parking space with a platform scheduling plan to obtain a matching result; license plate information can be accurately identified, visual management can be carried out on the park, and the platform management capability and the park operation efficiency are improved.
In a preferred embodiment of the present application, the process of identifying the license plate information includes:
splitting the video image into a plurality of image frames, carrying out gray processing on the image frames to obtain gray images, and carrying out noise reduction processing on the gray images;
if the color image is directly processed, the execution speed of the system is reduced, the storage space is increased, and reasonable graying is helpful to the extraction and subsequent processing of the image information, so that the storage space can be saved, and the processing speed is increased.
In a preferred embodiment of the present application, a Sobel edge detection operator is used to perform edge detection on the gray image after noise reduction, a maximum inter-class variance method is used to perform binarization processing on the gray image after edge detection to obtain a binarized image, the pixel value of the gray image after binarization processing is 0 or 255, and noise reduction processing is performed on the binarized image again;
the Sobel operator emphasizes the influence of the opposite side neighborhood pixels of the center pixel on the neighborhood pixels according to different weights of the distances between the neighborhood pixels and the current pixel, and weakens the effect of 4 opposite angle neighboring pixels; each pixel point in the image is convolved by the two kernels, one convolution checks that the response of the vertical edge of the image is maximum, the other convolution checks that the response of the vertical edge of the image is maximum, and the maximum value in the two convolutions is taken as the output value of the pixel point; therefore, the Sobel operator has a noise suppression effect, so that a plurality of isolated edge pixel points cannot appear.
In a preferred case of this embodiment, selecting the outline of the white connected region in the binary image by a rectangular bounding box to obtain a rectangular region to be determined, where the aspect ratio of the rectangular bounding box is a±a:b±b, A, B preset coefficients, a and B are preset error thresholds, and performing expansion and amplification on the rectangular region to be determined, if a plurality of rectangular regions to be determined exist, respectively calculating the pixel ratio of gray values of 255 and 0 in the rectangular region to be determined, and the frequency range of transition of gray in the rectangular region to be determined, if the pixel ratio of the rectangular region to be determined is lower than 0.25 and the frequency range of transition is located at [5,30], determining that the rectangular region to be determined is a license plate image, otherwise, continuing to select;
the length-width threshold value is taken to be 7 in consideration of the fact that in the license plate positioning process, the number plate information is reduced due to mathematical morphology operation of the number plate, the inclination of the number plate in the photographed license plate image is reduced, and therefore candidate areas with non-conforming length-width values are removed.
It is noted that, the license plate image is subjected to license plate character segmentation, two thresholds T1 and T2 are set, the license plate image is scanned from left to right, a first column with the pixel value of 255 being greater than the threshold T1 is obtained, namely a starting column of license plate Chinese characters, the starting column is marked as S, the license plate image is continuously scanned until a column with the pixel value of 255 being less than the threshold T1 is identified, the ending column of license plate Chinese characters is marked as O, if O-S is more than T2, O is the ending column of license plate Chinese characters, otherwise, the image is continuously scanned until a column with the width being greater than T2 from the S column is identified, and the number of pixels with the pixel value of 255 is less than the preset threshold is identified, and the column is the ending column of license plate Chinese characters;
scanning the rest characters continuously, obtaining columns with the pixel values of 255 and the pixel numbers of more than a threshold T1 as character starting columns, and repeating the steps by taking columns with the pixel values of 255 and the pixel numbers of less than the threshold T1 as character ending columns to finish character segmentation of license plates;
the first character of the license plate of China is Chinese characters, two thresholds are set according to the characteristics of the Chinese characters to divide the first Chinese character of the license plate, when the Chinese characters which are not communicated are divided, the improved method plays a significant role, the rest characters are English letters and Arabic numerals, the characters have no problem of non-connectivity, and therefore, the rest characters of the license plate can be divided by only one threshold.
In another preferable case of this embodiment, the cloud platform further includes a character library, the character library is generated based on big data, standard characters of various license plates are stored in the character library, the standard characters are selected, an outer edge line of the standard characters is drawn, a coordinate system is built by taking a geometric center point of the standard characters as an origin, a plurality of pairs of feature points on the outer edge line of the standard characters are selected as reference points, the reference points are marked as (Xi, yi) and (Mi, ni), and i= … n;
selecting the license plate character, drawing the outer edge line of the license plate character, establishing a coordinate system by taking the geometric center point of the character as an origin, and selecting a plurality of pairs of characteristic points on the outer edge line of the license plate character as detection points, wherein the detection points are marked as (Ui, vi) and (Zi, wi), and i= … n;
by the formulaObtaining a similarity value XS of license plate characters and standard characters, judging the license plate characters as corresponding standard characters if the similarity value XS is lower than a preset threshold value, obtaining license plate information, and if not, continuing to identify;
according to the application, the standard characters and the license plate characters are subjected to similarity comparison through the formulas by drawing the outer edge lines of the standard characters and the license plate characters and respectively selecting the reference points and the detection points, so that the license plate characters are identified.
In a preferred embodiment of the application, the gray scale image is divided into a plurality of overlapped image blocks, for each image block, image blocks around the image block are taken, norms between the image blocks are calculated and stored, and then non-dominant ordering is carried out, wherein the norms comprise MSE and MAE;
the existing noise reduction is mainly to calculate the similarity between image blocks obtained by Euclidean distance, but under the condition that the images are noisy, the noise can influence the calculation of the similarity between the image blocks, and the condition that the higher similarity is calculated between the image blocks which are originally dissimilar can possibly occur, so that the denoising effect of a follow-up image denoising algorithm can be influenced; according to the application, through non-dominant ordering of the two norms MSE and MAE on the image blocks, the best matched image block can be effectively found, so that the noise reduction efficiency of the image is improved.
In a preferred case of this embodiment, the specific calculation procedure of the norm is:
selecting image blocks and marking as P 1 Obtaining P 1 And marked as P 2 Block P of picture 1 Conversion to a formal representation of matrix i, converting image block P 2 Is converted into a form of matrix j, corresponding elements of the matrix i and the matrix j are subtracted, the sum of squares of the differences is calculated, and the sum is divided by the size of the corresponding matrix, thus obtaining an image block P 1 And image block P 2 Subtracting the corresponding elements of matrix i and matrix j and calculating the sum of the absolute values of the differences, dividing the sum by the size of the matrix to obtain the image block P 1 And adjacent image block P 2 MAE of (C);
the non-dominant ranking ranks individuals in the set by using the concept of Pareto optimal solution, and the higher the non-dominant state, the higher the individual hierarchy, the more abundant the data information it contains.
In another preferred case of the present embodiment, the top k image blocks P in the non-dominant order are sorted 2 Forming a matrix, and performing soft threshold contraction on singular values of the matrix by using different lambda values to obtain a low-rank matrix;
and acquiring singular values of the low-rank matrix, filtering smaller numerical values in the singular values, and merging the obtained noise-reduced image blocks into a noise-reduced gray image.
The foregoing describes one embodiment of the present application in detail, but the description is only a preferred embodiment of the present application and should not be construed as limiting the scope of the application. All equivalent changes and modifications within the scope of the present application are intended to be covered by the present application.
Claims (9)
1. A dock wagon operation behavior identification method for customs assistance management, comprising:
establishing a cloud platform, wherein the cloud platform is used for storing truck registration information and a platform scheduling plan;
acquiring a video image of a platform parking space, acquiring an occupied state of the platform parking space according to the video image, and uploading the occupied state to the cloud platform;
identifying an entrance truck, judging whether the truck is a registered truck, and distributing a parking space for the truck according to the platform dispatching plan and the occupied state;
and identifying license plate information of the truck occupying the parking space through the video image, judging whether the truck accords with the platform scheduling information, if so, carrying out platform operation, and if not, carrying out prompt.
2. A platform wagon operation behavior recognition method for customs assistance management according to claim 1, wherein the process of recognizing the license plate information includes:
splitting the video image into a plurality of image frames, carrying out gray processing on the image frames to obtain gray images, and carrying out noise reduction processing on the gray images.
3. The method for identifying the operation behavior of a platform truck for customs auxiliary management according to claim 2, wherein a Sobel edge detection operator is adopted to detect edges of the gray level image after noise reduction, a maximum inter-class variance method is adopted to perform binarization processing on the gray level image after edge detection to obtain a binarized image, the pixel value of the gray level image after binarization processing is 0 or 255, and noise reduction processing is performed on the binarized image again.
4. The method for identifying operation behaviors of a platform truck for customs auxiliary management according to claim 3, wherein a rectangular boundary box is used for selecting the outline of a white communication area in a binary image to obtain a rectangular area to be determined, the aspect ratio of the rectangular boundary box is A+/-a:B+/-B, A, B preset coefficients, a and B are preset error thresholds, the rectangular area to be determined is subjected to expansion amplification, if a plurality of rectangular areas to be determined exist, pixel ratios with gray values of 255 and 0 in the rectangular area to be determined are respectively calculated, and the jump frequency range of gray scales in the rectangular area to be determined is less than 0.25, if the pixel ratio of the rectangular area to be determined exists and the jump frequency range is in [5,30], otherwise, the rectangular area to be determined as a license plate image is determined, and the selection is continued.
5. The method for recognizing operation behaviors of a platform truck for customs auxiliary management according to claim 4, wherein license plate characters of the license plate image are segmented, two thresholds T1 and T2 are set, the license plate image is scanned from left to right, a first column with a pixel value of 255 and with the number of pixels larger than the threshold T1 is obtained, the first column is a starting column of license plate Chinese characters, the first column is marked as S, the license plate image is continuously scanned until a column with the pixel value of 255 and with the number of pixels smaller than the threshold T1 is recognized, the second column is marked as O, if O-S is larger than T2, O is the ending column of license plate Chinese characters, otherwise, the image is continuously scanned until a column with a width larger than T2 and the number of pixels with the pixel value of 255 is smaller than a preset threshold is recognized, and the first column is the ending column of license plate Chinese characters;
and continuing scanning the rest characters, obtaining a column with the pixel value of 255 and the pixel number of more than the threshold value T1 as a character starting column, and repeating the steps by taking a column with the pixel value of 255 and the pixel number of less than the threshold value T1 as a character ending column to finish character segmentation of the license plate.
6. The method for identifying the operation behavior of a platform truck for customs auxiliary management according to claim 5, wherein the cloud platform further comprises a character library, the character library is generated based on big data, standard characters with various license plates are stored, the standard characters are selected, the outer edge lines of the standard characters are drawn, a coordinate system is established by taking the geometric center point of the standard characters as an origin, a plurality of pairs of characteristic points on the outer edge lines of the standard characters are selected as reference points, and the reference points are marked as (Xi, yi) and (Mi, ni), and i= … n;
selecting the license plate character, drawing the outer edge line of the license plate character, establishing a coordinate system by taking the geometric center point of the character as an origin, and selecting a plurality of pairs of characteristic points on the outer edge line of the license plate character as detection points, wherein the detection points are marked as (Ui, vi) and (Zi, wi), and i= … n;
by the formulaObtaining a similarity value XS of license plate characters and standard characters, judging the license plate characters as corresponding standard characters if the similarity value XS is lower than a preset threshold value, obtaining license plate information, and otherwise, continuing to identify.
7. A platform wagon operation behavior recognition method for customs assistance management according to claim 2 or 3, wherein the noise reduction processing is as follows:
dividing the gray image into a plurality of overlapped image blocks, taking the image blocks around each image block, calculating and storing norms among the image blocks, and then carrying out non-dominant sorting, wherein the norms comprise MSE and MAE.
8. The method for identifying the operation behavior of a platform truck for customs assistance management according to claim 7, wherein the specific calculation process of the norm is:
selecting image blocks and marking as P 1 Obtaining P 1 And marked as P 2 Block P of picture 1 Conversion to a formal representation of matrix i, converting image block P 2 Conversion to a formal representation of matrix j, matrix i and matrix jSubtracting the corresponding elements and calculating the sum of the squares of the differences, dividing the sum by the size of the corresponding matrix to obtain an image block P 1 And image block P 2 Subtracting the corresponding elements of matrix i and matrix j and calculating the sum of the absolute values of the differences, dividing the sum by the size of the matrix to obtain the image block P 1 And adjacent image block P 2 Is a MAE of (c).
9. A platform wagon operation behavior recognition method for customs assistance management according to claim 7, wherein the top k image blocks P in the non-dominant order are sorted 2 Forming a matrix, and performing soft threshold contraction on singular values of the matrix by using different lambda values to obtain a low-rank matrix;
and acquiring singular values of the low-rank matrix, filtering smaller numerical values in the singular values, and merging the obtained noise-reduced image blocks into a noise-reduced gray image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310906974.5A CN117115701A (en) | 2023-07-24 | 2023-07-24 | Platform truck operation behavior identification method for customs auxiliary management |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310906974.5A CN117115701A (en) | 2023-07-24 | 2023-07-24 | Platform truck operation behavior identification method for customs auxiliary management |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117115701A true CN117115701A (en) | 2023-11-24 |
Family
ID=88811796
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310906974.5A Pending CN117115701A (en) | 2023-07-24 | 2023-07-24 | Platform truck operation behavior identification method for customs auxiliary management |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117115701A (en) |
-
2023
- 2023-07-24 CN CN202310906974.5A patent/CN117115701A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109886896B (en) | Blue license plate segmentation and correction method | |
| CN107729899B (en) | License plate number recognition method and device | |
| CN107045634B (en) | Text positioning method based on maximum stable extremum region and stroke width | |
| KR101403876B1 (en) | Method and Apparatus for Vehicle License Plate Recognition | |
| KR101795823B1 (en) | Text enhancement of a textual image undergoing optical character recognition | |
| JP5334042B2 (en) | Character string recognition method and character string recognition apparatus | |
| Parker et al. | An approach to license plate recognition | |
| CN108256521B (en) | Effective area positioning method for vehicle body color identification | |
| CN110197153B (en) | Automatic wall identification method in house type graph | |
| US9965695B1 (en) | Document image binarization method based on content type separation | |
| KR101182173B1 (en) | Method and system for recognizing vehicle plate | |
| CN108108734B (en) | License plate recognition method and device | |
| CN110210440B (en) | Table image layout analysis method and system | |
| CN101122953A (en) | Picture words segmentation method | |
| Khalifa et al. | Malaysian Vehicle License Plate Recognition. | |
| CN114529459A (en) | Method, system and medium for enhancing image edge | |
| He et al. | A new automatic extraction method of container identity codes | |
| Ahmed et al. | On-road automobile license plate recognition using co-occurrence matrix | |
| JP2903923B2 (en) | Car number recognition device | |
| JP4062987B2 (en) | Image area dividing method, image area dividing apparatus, and image area dividing program | |
| CN111242051A (en) | Vehicle identification optimization method and device and storage medium | |
| JPH08305795A (en) | Character recognizing method | |
| Polyakova et al. | Improvement of the Color Text Image Binarization Methodusing the Minimum-Distance Classifier | |
| CN117115701A (en) | Platform truck operation behavior identification method for customs auxiliary management | |
| CN114445814B (en) | Character region extraction method and computer-readable storage medium |
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 |