Automatic detection system and method for high-speed green channel

Abstract

The invention discloses an automatic detection system and method for a high-speed green channel, and relates to the technical field of automatic detection of the green channel, and the method comprises the steps: constructing a high-speed green channel automatic detection cloud platform, carrying out the image quality evaluation of an X-ray detection image, and obtaining a target detection image group; extracting a target X-ray detection image from the target detection image group, analyzing goods in the target X-ray detection image, and analyzing the similarity between the goods in the target X-ray detection image and goods in the goods comparison data to obtain goods type data; analyzing the cargo loading condition in the to-be-detected truck to obtain characteristic cargo loading data; the method comprises the following steps: acquiring characteristic cargo loading data of a to-be-detected truck, generating detection data of the to-be-detected truck based on the characteristic cargo loading data in combination with preset detection condition data in a cloud platform, displaying the detection data to high-speed workers through the cloud platform, and intelligently detecting the truck in a high-speed green channel.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic detection of green passes, and in particular to an automatic detection system and method for high-speed green passes. Background Art

[0002] Highway Green Pass refers to a convenient passage policy provided on highways to promote the transportation of agricultural products. It is mainly to support agricultural development, reduce transportation costs, and improve the efficiency of agricultural product circulation. Therefore, it is very important to conduct cargo inspection on vehicles in Highway Green Pass. Currently, it is mainly done manually to inspect the cargo of vehicles passing through the highway to check the type, quantity and loading status of the cargo. However, manual inspection of vehicles on the highway is very inefficient, and the results of manual inspections may be affected by the inspections and judgments of the inspectors, which is subjective and prone to omissions.

[0003] In addition, when inspecting vehicles on the highway green pass, X-ray equipment will also be used to identify the type and quantity of cargo in the vehicle. However, after using X-ray equipment to photograph the vehicle, professionals are still required to identify and interpret the X-ray images, which may cause misjudgment. If the X-ray equipment is used improperly, it will cause radiation to people and the surrounding environment, affecting people's health. Summary of the invention

[0004] The object of the present invention is to provide an automated detection system and method for high-speed green pass to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solution: an automated detection method for high-speed green pass, the method comprising:

[0006] Step S100: constructing a high-speed green pass automated detection cloud platform, obtaining X-ray detection images of the truck to be detected, performing image quality assessment on the X-ray detection images, and obtaining a target detection image group;

[0007] Step S200: extracting a target X-ray detection image from the target detection image group, obtaining cargo comparison data from the cloud platform, analyzing the cargo in the target X-ray detection image and the degree of similarity with the cargo in the cargo comparison data, and obtaining cargo type data;

[0008] Step S300: obtaining a target detection image group of the truck to be detected, analyzing the cargo loading situation in the truck to be detected based on the target detection image group and in combination with cargo type data, and obtaining characteristic cargo loading data;

[0009] Step S400: Acquire characteristic cargo loading data of the truck to be inspected, generate inspection data of the truck to be inspected based on the characteristic cargo loading data and the inspection condition data preset in the cloud platform, and display the inspection data to highway staff through the cloud platform to conduct intelligent inspection of trucks in the highway green channel.

[0010] Furthermore, step S100 includes:

[0011] Step S101: When the truck to be inspected enters the inspection channel in the highway, the inspection equipment uses X-rays to scan the cargo compartment of the truck to be inspected to obtain an X-ray inspection image;

[0012] Among them, a cold cathode X-ray tube is installed on the inspection equipment;

[0013] Step S102: using the image processing library to read the X-ray detection image, selecting the image portion corresponding to the cargo compartment, and recording it as a characteristic image region in the X-ray detection image;

[0014] Step S103: Calculate the regional characteristic brightness P of the characteristic image region of the X-ray detection image α :

[0015]

[0016] Where N represents the total number of pixels in the feature image area; G i Represented as the brightness value of the i-th pixel in the feature image area;

[0017] Step S104: removing the characteristic image region from the X-ray detection image to obtain a characteristic background image region of the X-ray detection image;

[0018] Step S105: Calculate the characteristic image quality S of the X-ray detection image:

[0019]

[0020] Among them, P σ It is expressed as the standard deviation of the pixels in the feature background image area;

[0021] Step S106: when the characteristic image quality of the X-ray detection image is greater than a preset characteristic image quality threshold, the X-ray detection image is saved and recorded as a target X-ray detection image; otherwise, the X-ray detection image is removed from each X-ray detection image of the truck to be detected;

[0022] Step S107: acquiring and collecting a plurality of target X-ray detection images of the truck to be detected to obtain a target detection image group;

[0023] In the above steps, a cold cathode X-ray tube is installed on the inspection equipment, because the cold cathode X-ray tube has higher brightness, faster response speed and longer service life than the hot cathode X-ray tube, thereby making the inspection equipment usable for a longer time and achieving better inspection results. In the subsequent process, the image quality of the acquired X-ray detection images is also evaluated, and X-ray detection images with unqualified image quality are screened out and eliminated, thereby avoiding deviations in subsequent results caused by image quality problems, thereby greatly improving the accuracy of subsequent cargo detection in trucks.

[0024] Further, step S200 includes:

[0025] Step S201: obtaining a target detection image group of the truck to be detected, obtaining a target X-ray detection image of the truck to be detected from the target detection image group, establishing a two-dimensional coordinate system with the target X-ray detection image as a unit length, and obtaining the coordinates of each pixel point in the target X-ray detection image;

[0026] Step S202: Gaussian filtering is performed on the target X-ray detection image. The specific formula is:

[0027]

[0028] Wherein, I(c,d) represents the brightness value of the pixel with coordinates (c,d) in the target X-ray detection image; σ represents the preset Gaussian standard deviation; w(c,d) represents the square of the distance between the pixel with coordinates (c,d) and each pixel in the preset area of ​​the target X-ray detection image; I′(c,d) represents the brightness value of the pixel with coordinates (c,d) after Gaussian filtering;

[0029] Step S203: using a preset contour extraction algorithm to extract contours of the cargo area in the target X-ray detection image, and setting an area threshold to remove cargo contours smaller than the area threshold, thereby obtaining a plurality of cargo contour images in the target X-ray detection image;

[0030] Step S204: performing image processing on the target X-ray detection image using histogram equalization, and obtaining cargo comparison data from the cloud platform, the cargo comparison data including standard comparison X-ray images of various cargo types;

[0031] Step S205: analyzing the similarity between the cargo contour image in the target X-ray detection image and the cargo of different cargo types in the cargo comparison data, wherein the similarity analysis process between the ath cargo contour image in the target X-ray detection image and the standard comparison X-ray image is to establish a two-dimensional coordinate system and obtain the coordinates of each pixel point of the cargo contour image and the standard comparison X-ray image;

[0032] Step S206: Use a feature point detection algorithm to extract the feature points of the a-th cargo contour image and the standard X-ray image, and record them as (x a ,y a ), (x′, y′), obtain the change matrix H of the a-th cargo contour image, and we can get:

[0033]

[0034] The change matrix H is optimized using the least square method, and the change matrix H is applied to the a-th cargo contour image to obtain the a-th cargo contour image after registration;

[0035] Step S207: Calculate the cargo approximation R between the ath cargo contour image and the bth standard control X-ray image in the cargo comparison data a,b :

[0036]

[0037] Among them, μ a It is represented as the average brightness value of the pixel points of the a-th cargo contour image; μ b It is expressed as the average brightness value of the pixels of the bth standard reference X-ray image; σ a 2 Expressed as the variance of the a-th cargo contour image; σ b 2 Expressed as the variance of the bth standard reference X-ray image; σ a,b It is represented as the covariance of the a-th cargo contour image and the b-th standard reference X-ray image; e1 and e2 are the preset first characteristic constant and second characteristic constant respectively;

[0038] Step S208: Obtain the a-th cargo contour image, and compare it with the maximum value of the cargo approximation values ​​of each standard control X-ray image in the cargo comparison data. When the maximum value is greater than the preset cargo approximation threshold, obtain the cargo type corresponding to the standard control X-ray image to which the maximum value belongs, and record it as the target cargo type. Obtain and aggregate several target cargo types of the target detection image group of the truck to be detected to obtain cargo type data of the truck to be detected.

[0039] Furthermore, step S300 includes:

[0040] Step S301: obtaining a target detection image group of the truck to be detected, constructing a three-dimensional image model of the cargo compartment of the truck to be detected based on the target X-ray detection image in the target detection image group, and establishing a three-dimensional coordinate system based on the three-dimensional image model;

[0041] Step S302: Set the unit length L, divide the three-dimensional image model into grids with a side length of the unit length, record the space formed by the grid as voxels, obtain the center point of the voxel, and emit a ray from the center point to any direction. When the number of intersections between the ray and the three-dimensional image model is an odd number, the voxel is determined to be filled, otherwise it is determined that the voxel is not filled;

[0042] Step S303: Obtain the filled voxels in the three-dimensional image model, and in combination with the cargo type data of the truck to be detected, obtain the total number of filled voxels corresponding to several target cargo types of the truck to be detected in the three-dimensional image model;

[0043] Step S304: Calculate the cargo loading rate F of a target cargo type = (L 3 ×M) / V sum , where M represents the total number of voxels filled in the 3D image model for a certain target cargo type, and V sum It is represented as the volume of the cargo compartment in the three-dimensional image model;

[0044] Step S305: Obtain cargo loading rates and total cargo loading rates of several target cargo types of the truck to be detected and aggregate them to obtain characteristic cargo loading data.

[0045] Furthermore, step S400 includes:

[0046] Step S401: Acquire characteristic cargo loading data of the truck to be inspected, obtain highway inspection condition data from the cloud platform, and make qualification determinations on several target cargo types in the truck to be inspected and the cargo loading rate of the truck to be inspected based on the inspection condition data, and generate inspection data of the truck to be inspected;

[0047] Step S402: Visualize the detection data generated by the cloud platform and display it to the highway staff who are inspecting the trucks to be inspected, so as to conduct intelligent inspection of the trucks in the highway green pass.

[0048] In order to better implement the above method, an automatic detection system for high-speed green pass is also provided in the package, the system includes a target detection image group module, a cargo type data module, a characteristic cargo loading data module, and an intelligent detection module;

[0049] The target detection image group module is used to perform image quality assessment on the X-ray detection image to obtain a target detection image group;

[0050] The cargo type data module is used to analyze the similarity between the cargo in the target X-ray detection image and the cargo in the cargo comparison data to obtain cargo type data;

[0051] The characteristic cargo loading data module is used to analyze the cargo loading situation in the truck to be inspected and obtain characteristic cargo loading data;

[0052] The intelligent detection module is used to generate detection data of the truck to be inspected based on the characteristic cargo loading data and the detection condition data preset in the cloud platform, and display the detection data to the highway staff through the cloud platform.

[0053] Further, the target detection image group module includes a feature image quality unit and a target detection image group unit;

[0054] A characteristic image quality unit, used for acquiring a regional characteristic brightness of a characteristic image region in an X-ray detection image, and calculating a characteristic image quality of the X-ray detection image;

[0055] The target detection image group unit is used to obtain and aggregate several target X-ray detection images of the truck to be detected according to the characteristic image quality of the X-ray detection image to obtain a target detection image group.

[0056] Further, the cargo type data module includes a cargo approximate value unit and a cargo type data unit;

[0057] A cargo approximation value unit, used to calculate the cargo contour image in the target X-ray detection image and the cargo approximation value of the standard comparison X-ray image in the cargo comparison data;

[0058] The cargo type data unit is used to acquire and collect cargo types of several items of the target detection image group of the truck to be detected, so as to obtain cargo type data of the truck to be detected.

[0059] Further, the characteristic cargo loading data module includes a cargo loading rate unit and a characteristic cargo loading data unit;

[0060] The cargo loading rate unit is used to calculate the cargo loading rates of several cargo types in the cargo type data;

[0061] The characteristic cargo loading data unit is used to obtain and collect the loading rates and total loading rates of several target cargo types of the truck to be inspected, so as to obtain characteristic cargo loading data.

[0062] Further, the intelligent detection module includes an intelligent detection unit;

[0063] The intelligent detection unit is used to obtain the characteristic cargo loading data of the truck to be inspected, obtain the inspection condition data of the highway from the cloud platform, generate the inspection data of the truck to be inspected, and perform intelligent inspection on the trucks in the highway green channel.

[0064] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention realizes the automated detection of the green channel in the expressway company, firstly, the acquired X-ray detection images are screened, so that the image quality for analyzing the cargo in the truck is guaranteed, and the accuracy of the analysis is fundamentally improved; secondly, according to the cargo contour image in the X-ray detection image, the approximation degree is analyzed with the standard control X-ray image, the target cargo type in the X-ray detection image is obtained, and a three-dimensional model is constructed to obtain the cargo loading rate of the target cargo type, and then according to the detection condition data, the cargo loading status of the truck is intelligently detected to determine whether the cargo type and loading rate in the truck meet the requirements, which not only greatly improves the detection efficiency, but also significantly improves the accuracy of cargo detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 It is a method flow chart of an automated detection system and method for high-speed green pass of the present invention;

[0066] Figure 2 It is a module schematic diagram of an automated detection system and method for high-speed green pass of the present invention. DETAILED DESCRIPTION

[0067] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0068] Example: Figure 1-Figure 2 As shown, the present invention provides a technical solution, an automated detection method for high-speed green pass, the method comprising:

[0069] Step S100: constructing a high-speed green pass automated detection cloud platform, obtaining X-ray detection images of the truck to be detected, performing image quality assessment on the X-ray detection images, and obtaining a target detection image group;

[0070] Wherein, step S100 includes:

[0071] Step S101: When the truck to be inspected enters the inspection channel in the highway, the inspection equipment uses X-rays to scan the cargo compartment of the truck to be inspected to obtain an X-ray inspection image;

[0072] Among them, a cold cathode X-ray tube is installed on the inspection equipment;

[0073] Step S102: using the image processing library to read the X-ray detection image, selecting the image portion corresponding to the cargo compartment, and recording it as a characteristic image region in the X-ray detection image;

[0074] Step S103: Calculate the regional characteristic brightness P of the characteristic image region of the X-ray detection image α :

[0075]

[0076] Where N represents the total number of pixels in the feature image area; G i Represented as the brightness value of the i-th pixel in the feature image area;

[0077] Step S104: removing the characteristic image region from the X-ray detection image to obtain a characteristic background image region of the X-ray detection image;

[0078] Step S105: Calculate the characteristic image quality S of the X-ray detection image:

[0079]

[0080] Among them, P σ It is expressed as the standard deviation of the pixels in the feature background image area;

[0081] Step S106: when the characteristic image quality of the X-ray detection image is greater than a preset characteristic image quality threshold, the X-ray detection image is saved and recorded as a target X-ray detection image; otherwise, the X-ray detection image is removed from each X-ray detection image of the truck to be detected;

[0082] Step S107: acquiring and collecting a plurality of target X-ray detection images of the truck to be detected to obtain a target detection image group;

[0083] Step S200: extracting a target X-ray detection image from the target detection image group, obtaining cargo comparison data from the cloud platform, analyzing the cargo in the target X-ray detection image and the degree of similarity with the cargo in the cargo comparison data, and obtaining cargo type data;

[0084] Wherein, step S200 includes:

[0085] Step S201: obtaining a target detection image group of the truck to be detected, obtaining a target X-ray detection image of the truck to be detected from the target detection image group, establishing a two-dimensional coordinate system with the target X-ray detection image as a unit length, and obtaining the coordinates of each pixel point in the target X-ray detection image;

[0086] Step S202: Gaussian filtering is performed on the target X-ray detection image. The specific formula is:

[0087]

[0088] Wherein, I(c,d) represents the brightness value of the pixel with coordinates (c,d) in the target X-ray detection image; σ represents the preset Gaussian standard deviation; w(c,d) represents the square of the distance between the pixel with coordinates (c,d) and each pixel in the preset area of ​​the target X-ray detection image; I′(c,d) represents the brightness value of the pixel with coordinates (c,d) after Gaussian filtering;

[0089] Step S203: using a preset contour extraction algorithm to extract contours of the cargo area in the target X-ray detection image, and setting an area threshold to remove cargo contours smaller than the area threshold, thereby obtaining a plurality of cargo contour images in the target X-ray detection image;

[0090] Step S204: performing image processing on the target X-ray detection image using histogram equalization, and obtaining cargo comparison data from the cloud platform, the cargo comparison data including standard comparison X-ray images of various cargo types;

[0091] For example, the various types of goods include carrots, potatoes, etc.;

[0092] Step S205: analyzing the similarity between the cargo contour image in the target X-ray detection image and the cargo of different cargo types in the cargo comparison data, wherein the similarity analysis process between the ath cargo contour image in the target X-ray detection image and the standard comparison X-ray image is to establish a two-dimensional coordinate system and obtain the coordinates of each pixel point of the cargo contour image and the standard comparison X-ray image;

[0093] Step S206: Use a feature point detection algorithm to extract the feature points of the a-th cargo contour image and the standard X-ray image, and record them as (x a ,y a ), (x′, y′), obtain the change matrix H of the a-th cargo contour image, and we can get:

[0094]

[0095] The change matrix H is optimized using the least square method, and the change matrix H is applied to the a-th cargo contour image to obtain the a-th cargo contour image after registration;

[0096] For example, feature point detection algorithms include SIFT, SURF, ORB, etc.

[0097] Step S207: Calculate the cargo approximation R between the ath cargo contour image and the bth standard control X-ray image in the cargo comparison data a,b:

[0098]

[0099] Among them, μ a It is represented as the average brightness value of the pixel points of the a-th cargo contour image; μ b It is expressed as the average brightness value of the pixels of the bth standard reference X-ray image; σ a 2 Expressed as the variance of the a-th cargo contour image; σ b 2 Expressed as the variance of the bth standard reference X-ray image; σ a,b It is represented as the covariance of the a-th cargo contour image and the b-th standard reference X-ray image; e1 and e2 are the preset first characteristic constant and second characteristic constant respectively;

[0100] For example, the average value μ1 of the brightness value of the pixels of the first cargo contour image is expressed as 200; the average value μ2 of the brightness value of the pixels of the second standard reference X-ray image is expressed as 230; the variance σ1 of the first cargo contour image is expressed as 2 Expressed as 150; the variance of the second standard reference X-ray image σ2 2 Expressed as 180; the covariance σ between the a-th cargo contour image and the b-th standard reference X-ray image 1,2 It is represented as 130; the preset first characteristic constant and second characteristic constant e1 and e2 are 1 and 1 respectively;

[0101] Calculate the cargo approximation R between the first cargo contour image and the second standard control X-ray image in the cargo comparison data 1,2 :

[0102]

[0103] Step S208: obtaining the maximum value of the cargo approximation values ​​of the a-th cargo contour image and each standard control X-ray image in the cargo comparison data; when the maximum value is greater than a preset cargo approximation threshold, obtaining the cargo type corresponding to the standard control X-ray image to which the maximum value belongs, and recording it as the target cargo type; obtaining and aggregating the target cargo types of several items of the target detection image group of the truck to be detected, and obtaining cargo type data of the truck to be detected;

[0104] Step S300: obtaining a target detection image group of the truck to be detected, analyzing the cargo loading situation in the truck to be detected based on the target detection image group and in combination with cargo type data, and obtaining characteristic cargo loading data;

[0105] Wherein, step S300 includes:

[0106] Step S301: obtaining a target detection image group of the truck to be detected, constructing a three-dimensional image model of the cargo compartment of the truck to be detected based on the target X-ray detection image in the target detection image group, and establishing a three-dimensional coordinate system based on the three-dimensional image model;

[0107] Step S302: Set the unit length L, divide the three-dimensional image model into grids with a side length of the unit length, record the space formed by the grid as voxels, obtain the center point of the voxel, and emit a ray from the center point to any direction. When the number of intersections between the ray and the three-dimensional image model is an odd number, the voxel is determined to be filled, otherwise it is determined that the voxel is not filled;

[0108] Step S303: Obtain the filled voxels in the three-dimensional image model, and in combination with the cargo type data of the truck to be detected, obtain the total number of filled voxels corresponding to several target cargo types of the truck to be detected in the three-dimensional image model;

[0109] Step S304: Calculate the cargo loading rate F of a target cargo type = (L 3 ×M) / V sum , where M represents the total number of voxels filled in the 3D image model for a certain target cargo type, and V sum It is represented as the volume of the cargo compartment in the three-dimensional image model;

[0110] Step S305: obtaining cargo loading rates and total cargo loading rates of several cargo types of the truck to be detected and aggregating them to obtain characteristic cargo loading data;

[0111] Step S400: Acquire characteristic cargo loading data of the truck to be detected, generate detection data of the truck to be detected based on the characteristic cargo loading data and the detection condition data preset in the cloud platform, and display the detection data to the highway staff through the cloud platform to perform intelligent detection on the trucks in the highway green pass;

[0112] Wherein, step S400 includes:

[0113] Step S401: Acquire characteristic cargo loading data of the truck to be inspected, obtain highway inspection condition data from the cloud platform, and make qualification determinations on several target cargo types in the truck to be inspected and the cargo loading rate of the truck to be inspected based on the inspection condition data, and generate inspection data of the truck to be inspected;

[0114] Step S402: Visualize the detection data generated by the cloud platform and display it to the highway staff who are testing the trucks to be tested, and perform intelligent detection on the trucks in the highway green pass;

[0115] In order to better implement the above method, an automatic detection system for high-speed green pass is also provided in the package, the system includes a target detection image group module, a cargo type data module, a characteristic cargo loading data module, and an intelligent detection module;

[0116] The target detection image group module is used to perform image quality assessment on the X-ray detection image to obtain a target detection image group;

[0117] The cargo type data module is used to analyze the similarity between the cargo in the target X-ray detection image and the cargo in the cargo comparison data to obtain cargo type data;

[0118] The characteristic cargo loading data module is used to analyze the cargo loading situation in the truck to be inspected and obtain characteristic cargo loading data;

[0119] The intelligent detection module is used to generate the detection data of the truck to be detected based on the characteristic cargo loading data and the detection condition data preset in the cloud platform, and display the detection data to the highway staff through the cloud platform;

[0120] Wherein, the target detection image group module includes a feature image quality unit and a target detection image group unit;

[0121] A characteristic image quality unit, used for acquiring a regional characteristic brightness of a characteristic image region in an X-ray detection image, and calculating a characteristic image quality of the X-ray detection image;

[0122] The target detection image group unit is used to obtain and aggregate a plurality of target X-ray detection images of the truck to be detected according to the characteristic image quality of the X-ray detection image to obtain a target detection image group;

[0123] The cargo type data module includes a cargo approximate value unit and a cargo type data unit;

[0124] A cargo approximation value unit, used to calculate the cargo contour image in the target X-ray detection image and the cargo approximation value of the standard comparison X-ray image in the cargo comparison data;

[0125] A cargo type data unit is used to obtain and collect cargo types of several items of the target detection image group of the truck to be detected, so as to obtain cargo type data of the truck to be detected;

[0126] Among them, the characteristic cargo loading data module includes a cargo loading rate unit and a characteristic cargo loading data unit;

[0127] The cargo loading rate unit is used to calculate the cargo loading rates of several cargo types in the cargo type data;

[0128] The characteristic cargo loading data unit is used to obtain and collect the loading rates and total loading rates of several target cargo types of the truck to be inspected, and obtain characteristic cargo loading data;

[0129] Wherein, the intelligent detection module includes an intelligent detection unit;

[0130] The intelligent detection unit is used to obtain the characteristic cargo loading data of the truck to be inspected, obtain the inspection condition data of the highway from the cloud platform, generate the inspection data of the truck to be inspected, and perform intelligent inspection on the trucks in the highway green channel.

[0131] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. An automated detection method for high-speed green pass, characterized in that: The method comprises: Step S100: constructing a high-speed green pass automated detection cloud platform, obtaining an X-ray detection image of the truck to be detected, performing image quality assessment on the X-ray detection image, and obtaining a target detection image group; Step S200: extracting a target X-ray detection image from the target detection image group, acquiring cargo comparison data from the cloud platform, analyzing the cargo in the target X-ray detection image and the degree of similarity with the cargo in the cargo comparison data, and obtaining cargo type data; Step S300: acquiring a target detection image group of the truck to be detected, analyzing the cargo loading condition in the truck to be detected based on the target detection image group and in combination with the cargo type data, and obtaining characteristic cargo loading data; Step S400: Acquire characteristic cargo loading data of the truck to be inspected, generate inspection data of the truck to be inspected based on the characteristic cargo loading data and the inspection condition data preset in the cloud platform, and display the inspection data to highway staff through the cloud platform to perform intelligent inspection on trucks in the highway green channel.

2. The automated detection method for high-speed green pass according to claim 1, characterized in that: The step S100 includes: Step S101: When a truck to be inspected enters an inspection channel in a highway, an inspection device uses X-rays to scan the cargo compartment of the truck to be inspected to obtain an X-ray inspection image; Step S102: using an image processing library to read the X-ray detection image, selecting an image portion corresponding to the cargo compartment, and recording it as a characteristic image region in the X-ray detection image; Step S103: Calculate the regional characteristic brightness P of the characteristic image region of the X-ray detection image α : Wherein, N represents the total number of pixels in the feature image area; G i It is represented as the brightness value of the i-th pixel in the feature image area; Step S104: removing the characteristic image region from the X-ray detection image to obtain a characteristic background image region of the X-ray detection image; Step S105: Calculate the characteristic image quality S of the X-ray detection image: Among them, P σ It is expressed as the standard deviation of the pixels in the characteristic background image area; Step S106: when the characteristic image quality of the X-ray detection image is greater than a preset characteristic image quality threshold, the X-ray detection image is saved and recorded as a target X-ray detection image; otherwise, the X-ray detection image is removed from each X-ray detection image of the truck to be detected; Step S107: acquiring and aggregating a plurality of target X-ray detection images of the truck to be detected to obtain a target detection image group.

3. The automated detection method for high-speed green pass according to claim 2, characterized in that: The step S200 includes: Step S201: obtaining a target detection image group of the truck to be detected, obtaining a target X-ray detection image of the truck to be detected from the target detection image group, establishing a two-dimensional coordinate system with the target X-ray detection image as a unit length, and obtaining the coordinates of each pixel point in the target X-ray detection image; Step S202: Gaussian filtering is performed on the target X-ray detection image. The specific formula is: Wherein, I(c, d) represents the brightness value of the pixel with coordinates (c, d) in the target X-ray detection image; σ represents the preset Gaussian standard deviation; w(c, d) represents the square of the distance between the pixel with coordinates (c, d) and each pixel in the preset area of ​​the target X-ray detection image; I′(c, d) represents the brightness value of the pixel with coordinates (c, d) after Gaussian filtering; Step S203: using a preset contour extraction algorithm to extract contours of the cargo area in the target X-ray detection image, and setting an area threshold to remove cargo contours smaller than the area threshold, thereby obtaining a plurality of cargo contour images in the target X-ray detection image; Step S204: performing image processing on the target X-ray detection image using histogram equalization, and obtaining cargo comparison data from the cloud platform, wherein the cargo comparison data includes standard comparison X-ray images of various cargo types; Step S205: analyzing the similarity between the cargo contour image in the target X-ray detection image and the cargo of different cargo types in the cargo comparison data, wherein the similarity analysis process between the ath cargo contour image in the target X-ray detection image and the standard comparison X-ray image is to establish a two-dimensional coordinate system and obtain the coordinates of each pixel point of the cargo contour image and the standard comparison X-ray image; Step S206: Use a feature point detection algorithm to extract the feature points of the ath cargo contour image and the standard X-ray image, and record them as (x a ,y a ), (x′, y′), obtain the change matrix H of the a-th cargo contour image, and we can get: Optimizing the change matrix H using the least square method, applying the change matrix H to the a-th cargo contour image, and obtaining the a-th cargo contour image after registration; Step S207: Calculate the cargo approximation R between the ath cargo contour image and the bth standard control X-ray image in the cargo control data. a,b : Among them, μ a It is represented as the average brightness value of the pixel points of the a-th cargo contour image; μ b It is represented by the average brightness value of the pixels of the bth standard reference X-ray image; σ a 2 Expressed as the variance of the ath cargo contour image; σ b 2 Expressed as the variance of the bth standard reference X-ray image; σ a,b It is represented as the covariance of the a-th cargo contour image and the b-th standard reference X-ray image; e1 and e2 are respectively the preset first characteristic constant and the second characteristic constant; Step S208: Obtain the a-th cargo contour image and the maximum value of the cargo approximation values ​​of each standard control X-ray image in the cargo comparison data; when the maximum value is greater than a preset cargo approximation threshold, obtain the cargo type corresponding to the standard control X-ray image to which the maximum value belongs, and record it as the target cargo type; obtain and aggregate the target cargo types of several items of the target detection image group of the truck to be detected, and obtain the cargo type data of the truck to be detected.

4. The automated detection method for high-speed green pass according to claim 3, characterized in that: The step S300 includes: Step S301: acquiring a target detection image group of the truck to be detected, constructing a three-dimensional image model of the cargo compartment of the truck to be detected based on the target X-ray detection image in the target detection image group, and establishing a three-dimensional coordinate system based on the three-dimensional image model; Step S302: setting a unit length L, dividing the three-dimensional image model into grids with a side length of the unit length, recording the space formed by the grid as a voxel, obtaining the center point of the voxel, emitting a ray from the center point in any direction, and determining that the voxel is filled when the number of intersections between the ray and the three-dimensional image model is an odd number, otherwise determining that the voxel is not filled; Step S303: obtaining the filled voxels in the three-dimensional image model, and combining the cargo type data of the truck to be detected, obtaining the total number of filled voxels corresponding to several target cargo types of the truck to be detected in the three-dimensional image model; Step S304: Calculate the cargo loading rate F of a target cargo type = (L 3 ×M) / V sum , where M represents the total number of voxels filled in the three-dimensional image model for a certain target cargo type, and V sum Represented as the volume of the cargo compartment in the three-dimensional image model; Step S305: Obtain cargo loading rates and total cargo loading rates of several target cargo types of the truck to be detected and aggregate them to obtain characteristic cargo loading data.

5. The automated detection method for high-speed green pass according to claim 4, characterized in that: The step S400 includes: Step S401: Acquire characteristic cargo loading data of the truck to be detected, acquire detection condition data of the highway from the cloud platform, perform qualification determination on several target cargo types in the truck to be detected and the cargo loading rate of the truck to be detected based on the detection condition data, and generate detection data of the truck to be detected; Step S402: Visualize the detection data generated by the cloud platform and display it to the highway staff who are inspecting the trucks to be inspected, so as to perform intelligent inspection on the trucks in the highway green pass.

6. An automated detection system for high-speed green pass, used to execute an automated detection method for high-speed green pass as described in any one of claims 1 to 5, characterized in that: The system includes a target detection image group module, a cargo type data module, a characteristic cargo loading data module, and an intelligent detection module; The target detection image group module is used to perform image quality assessment on the X-ray detection image to obtain a target detection image group; The cargo type data module is used to analyze the similarity between the cargo in the target X-ray detection image and the cargo in the cargo comparison data to obtain cargo type data; The characteristic cargo loading data module is used to analyze the cargo loading situation in the truck to be inspected and obtain characteristic cargo loading data; The intelligent detection module is used to generate the detection data of the truck to be detected based on the characteristic cargo loading data and the detection condition data preset in the cloud platform, and display the detection data to the highway staff through the cloud platform.

7. The automatic detection system for high-speed green pass according to claim 6, characterized in that: The target detection image group module includes a feature image quality unit and a target detection image group unit; The characteristic image quality unit is used to obtain the regional characteristic brightness of the characteristic image area in the X-ray detection image and calculate the characteristic image quality of the X-ray detection image; The target detection image group unit is used to obtain and aggregate a plurality of target X-ray detection images of the truck to be detected according to the characteristic image quality of the X-ray detection image to obtain a target detection image group.

8. The automatic detection system for high-speed green pass according to claim 6, characterized in that: The cargo type data module includes a cargo approximate value unit and a cargo type data unit; The cargo approximation value unit is used to calculate the cargo approximation value of the cargo contour image in the target X-ray detection image and the standard comparison X-ray image in the cargo comparison data; The cargo type data unit is used to acquire and collect cargo types of several items of the target detection image group of the truck to be detected, so as to obtain cargo type data of the truck to be detected.

9. The automatic detection system for high-speed green pass according to claim 6, characterized in that: The characteristic cargo loading data module includes a cargo loading rate unit and a characteristic cargo loading data unit; The cargo loading rate unit is used to calculate the cargo loading rates of several target cargo types in the cargo type data; The characteristic cargo loading data unit is used to obtain and collect the loading rates and total loading rates of several target cargo types of the truck to be detected to obtain characteristic cargo loading data.

10. The automatic detection system for high-speed green pass according to claim 6, characterized in that: The intelligent detection module includes an intelligent detection unit; The intelligent detection unit is used to obtain characteristic cargo loading data of the truck to be detected, obtain detection condition data of the highway from the cloud platform, generate detection data of the truck to be detected, and perform intelligent detection on trucks within the highway green pass.

Images