Intelligent sorting system, method and storage medium for express parcels based on X-ray images
By using data processing terminals and learning classifiers to extract and classify X-ray images in the express parcel detection system, the problems of low manual detection efficiency and poor accuracy in the prior art are solved, and efficient and accurate violations of express parcels are achieved Product identification and sorting.
Patent Information
- Application Number
- CN201910687372.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2019-07-26
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2039-07-26
AI Technical Summary
In the prior art, the detection of contraband products of express parcels relies on manual observation of X-ray images, resulting in low efficiency, poor accuracy, consumes a lot of manpower and material resources, and is prone to missed inspections and missed inspections.
Design an intelligent sorting system based on X-ray images, through data processing terminal, transmission device, scanning and detection device and sorting device, use preset detection algorithm to detect express parcels, extract feature information, and build a learning classifier , realize the accurate identification and sorting of contraband.
It improves the inspection efficiency and accuracy of express parcels, realizes accurate identification and sorting of contraband products, and reduces security inspection costs.
Smart Images

Figure CN110711718B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of automatic control technology, and in particular to an intelligent sorting system, method and storage medium for express parcels based on X-ray images. Background Art
[0002] In recent years, thanks to the rapid development of e-commerce, the express delivery industry has experienced explosive growth. In particular, from 2011 to 2015, the express delivery industry experienced an average growth rate of more than 50%. The rapid development of the industry has also caused tremendous pressure on express delivery security inspection. Before delivery, express parcels need to be pre-checked to see if there are any contraband in the parcels, because the authenticity of the waybill information on the express parcels cannot be reliably guaranteed.
[0003] For safety reasons, traditionally, professional security inspectors are required to manually observe the X-ray images of each piece of cargo next to the X-ray machine to confirm whether the cargo is prohibited. This requires a huge amount of manpower and material resources, and work efficiency decreases as the fatigue of the security inspectors increases, and the number of missed inspections and wrong inspections will also increase accordingly.
[0004] At present, the main method of identifying prohibited items in express delivery is to rely on manual security inspection, which usually requires several groups of staff to work in shifts. Each group includes at least one professional security inspector and two sorting personnel, which requires a huge amount of manpower and material resources. Summary of the invention
[0005] The technical problem to be solved by the present invention is to provide an intelligent sorting system, method and storage medium for express parcels based on X-ray images, which can improve the inspection efficiency and accuracy of express parcels, and accurately identify and sort prohibited items.
[0006] In a first aspect, an embodiment of the present invention provides an intelligent sorting system for express parcels based on X-ray images, comprising a data processing terminal, a transmission device, a scanning and detection device, and a sorting device;
[0007] The express parcel is conveyed by the conveying device, passes through the scanning and detecting device, and then arrives at the sorting device;
[0008] The scanning and detection device is used to collect the location information, serial number information, and X-ray image of the express package and transmit them to the data processing terminal;
[0009] The data processing terminal is used to detect the express parcel according to a preset detection algorithm, and when it is determined that the express parcel contains contraband, write the X-ray image into a contraband category label, display the detection result and issue an alarm; and send a sorting instruction carrying the location information and the serial number information to the sorting device;
[0010] The sorting device is used to sort out the express parcels according to the sorting instructions.
[0011] Preferably, the preset detection algorithm includes:
[0012] 1) Establish an X-ray image database of express parcels, divide them into two categories: prohibited items and non-prohibited items, and set the prohibited items category as the target detection category;
[0013] 2) Divide the X-ray image into multiple connected regions, each of which is a 3*3 pixel block.
[0014] 3) Extract the features of each connected area. Specifically,
[0015] A. Use the following formula to calculate the local rotation-invariant and uniform-mode characteristic variables of the connected region:
[0016]
[0017] B. Set the transformation threshold of the feature variable parameters obtained in step A above to N times, and add N-dimensional features on the basis of the feature variable dimension. The formula is:
[0018] Among them, P is the number of pixels in the neighborhood, R is the neighborhood radius, and g p is the gray value of the neighborhood pixel, g c is the grayscale pixel value of the center point, S(x) is the sign function, and U(x) calculates the number of 0 / 1 transitions in the local binary pattern;
[0019] C. Select a gradient operator to calculate the directional gradient density feature histogram of the connected area:
[0020]
[0021] Where M(x,y) is the gradient magnitude of the pixel, θ(x,y) is the gradient direction of the pixel, and θ(x,y) is divided into several bins to obtain the directional gradient density feature histogram;
[0022] D. Combine the feature variables obtained in the above steps to construct a feature descriptor:
[0023] feature=concat(feature1,feature2,feature3,...,featureL)
[0024] 4) Construct a learning classifier, train the features represented by the feature descriptor, and learn the feature classification decision surface y=wx+b;
[0025] A. The dimension of the input vector is the dimension of the features represented by the feature descriptor;
[0026] B. Normalize the scale of feature variables:
[0027]
[0028] C. Design and use an enhanced kernel function based on Manhattan distance and Euclidean distance improvement, the formula is:
[0029]
[0030] The X-ray images of express packages are used as training data to train the learning classifier designed above to learn the feature classification decision surface;
[0031] 5) Access the real-time image data from the imaging system and perform cutting processing on the image based on the outline of the express package;
[0032] 6) Construct a feature descriptor for the cut express package image and put it into the trained learning model to obtain the prohibited item label information of the express package.
[0033] Preferably, the conveying device includes a driving motor and a conveyor belt driven by the driving motor; the sorting device includes a sorting controller, a robot and a sorting line.
[0034] Preferably, the scanning and detection device includes an X-ray security inspection device, and a first detection device and a second detection device respectively installed at both ends of the X-ray security inspection device along the direction of the conveyor belt; the first detector device includes a first position sensor, a first data collector and a first scanning controller, the second detector device includes a second position sensor, a second data collector and a second scanning controller, and the X-ray security inspection equipment includes a third data collector; the sorting device includes a sorting controller, a manipulator and a sorting line.
[0035] In a second aspect, an embodiment of the present invention further provides an intelligent sorting method for express parcels based on X-ray images, which is applicable to a data processing terminal, comprising:
[0036] Receiving the location information, serial number information, and X-ray image of the express package transmitted by the scanning and detection device;
[0037] The express package is inspected according to a preset inspection algorithm. When it is determined that the express package contains contraband, the X-ray image is written into a contraband category label, the inspection result is displayed, and an alarm is issued;
[0038] A sorting instruction carrying the location information and the serial number information is sent to the sorting device, so that the express parcel is sorted out according to the sorting instruction.
[0039] Preferably, the preset detection algorithm includes:
[0040] 1) Establish an X-ray image database of express parcels, divide them into two categories: prohibited items and non-prohibited items, and set the prohibited items category as the target detection category;
[0041] 2) Divide the X-ray image into multiple connected regions, each of which is a 3*3 pixel block.
[0042] 3) Extract the features of each connected area. Specifically,
[0043] A. Use the following formula to calculate the local rotation-invariant and uniform-mode characteristic variables of the connected region:
[0044]
[0045] B. Set the transformation threshold of the feature variable parameters obtained in step A above to N times, and add N-dimensional features on the basis of the feature variable dimension. The formula is:
[0046] Among them, P is the number of pixels in the neighborhood, R is the neighborhood radius, and g p is the gray value of the neighborhood pixel, g c is the grayscale pixel value of the center point, S(x) is the sign function, and U(x) calculates the number of 0 / 1 transitions in the local binary pattern;
[0047] C. Select a gradient operator to calculate the directional gradient density feature histogram of the connected area:
[0048]
[0049] Where M(x,y) is the gradient magnitude of the pixel, θ(x,y) is the gradient direction of the pixel, and θ(x,y) is divided into several bins to obtain the directional gradient density feature histogram;
[0050] D. Combine the feature variables obtained in the above steps to construct a feature descriptor:
[0051] feature=concat(feature1,feature2,feature3,...,featureL)
[0052] 4) Construct a learning classifier, train the features represented by the feature descriptor, and learn the feature classification decision surface y=wx+b;
[0053] A. The dimension of the input vector is the dimension of the features represented by the feature descriptor;
[0054] B. Normalize the scale of feature variables:
[0055]
[0056] C. Design and use an enhanced kernel function based on Manhattan distance and Euclidean distance improvement, the formula is:
[0057]
[0058] The X-ray images of express packages are used as training data to train the learning classifier designed above to learn the feature classification decision surface;
[0059] 5) Access the real-time image data from the imaging system and perform cutting processing on the image based on the outline of the express package;
[0060] 6) Construct a feature descriptor for the cut express package image and put it into the trained learning model to obtain the prohibited item label information of the express package.
[0061] Preferably, the scanning and detection device includes an X-ray security inspection device, and a first detection device and a second detection device respectively installed at both ends of the X-ray security inspection device along the direction of the conveyor belt; the first detector device includes a first position sensor, a first data collector and a first scanning controller, the second detector device includes a second position sensor, a second data collector and a second scanning controller, and the X-ray security inspection equipment includes a third data collector; the sorting device includes a sorting controller, a manipulator and a sorting line.
[0062] In a third aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the storage medium includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the above-mentioned X-ray image-based intelligent sorting method for express parcels.
[0063] Implementing the embodiments of the present invention has the following beneficial effects:
[0064] The present invention provides an intelligent sorting system and method for express parcels based on X-ray images. The express parcels are transmitted by a transmission device, and after passing through a scanning and detection device, they arrive at a sorting device; the scanning and detection device is used to transmit the location information, serial number information, and X-ray image of the collected express parcels to the data processing terminal; the data processing terminal is used to detect the express parcel according to a preset detection algorithm, and when it is determined that the express parcel contains contraband, the X-ray image is written into the contraband category label, the detection result is displayed and an alarm is issued; and a sorting instruction carrying the location information and the serial number information is issued to the sorting device; the sorting device is used to sort out the express parcel according to the sorting instruction. The present invention can improve the inspection efficiency and accuracy of express parcels, and accurately identify and sort contraband. BRIEF DESCRIPTION OF THE DRAWINGS
[0065] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the implementation mode will be briefly introduced below. Obviously, the drawings described below are only some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0066] Figure 1 It is a structural schematic diagram of an express parcel intelligent sorting system based on X-ray images provided by an embodiment of the present invention;
[0067] Figure 2 is another structural schematic diagram of an express parcel intelligent sorting system based on X-ray images provided by one embodiment of the present invention;
[0068] Figure 3 It is another structural schematic diagram of an express parcel intelligent sorting system based on X-ray images provided by an embodiment of the present invention;
[0069] Figure 4 It is a flow chart of an intelligent express parcel sorting method based on X-ray images provided in one embodiment of the present invention. DETAILED DESCRIPTION
[0070] 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.
[0071] It should be understood that the step numbers used in this document are only for convenience of description and are not intended to limit the order in which the steps are executed.
[0072] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.
[0073] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.
[0074] The term "and / or" means and includes any and all possible combinations of one or more of the associated listed items.
[0075] The first embodiment of the present invention:
[0076] See also Figure 1-3 , an intelligent sorting system for express parcels based on X-ray images, comprising a data processing terminal, a conveying device, a scanning and detecting device and a sorting device;
[0077] The express parcel is conveyed by the conveying device, passes through the scanning and detecting device, and then arrives at the sorting device;
[0078] The scanning and detection device is used to collect the location information, serial number information, and X-ray image of the express package and transmit them to the data processing terminal;
[0079] The data processing terminal is used to detect the express parcel according to a preset detection algorithm, and when it is determined that the express parcel contains contraband, write the X-ray image into a contraband category label, display the detection result and issue an alarm; and send a sorting instruction carrying the location information and the serial number information to the sorting device;
[0080] The sorting device is used to sort out the express parcels according to the sorting instructions.
[0081] Wherein, the conveying device includes a driving motor and a conveyor belt driven by the driving motor; the sorting device includes a sorting controller, a robot and a sorting line.
[0082] Among them, the scanning and detection device includes an X-ray security inspection device, and a first detection device and a second detection device respectively installed at both ends of the X-ray security inspection device along the direction of the conveyor belt; the first detector device includes a first position sensor, a first data collector and a first scanning controller, the second detector device includes a second position sensor, a second data collector and a second scanning controller, and the X-ray security inspection equipment includes a third data collector; the sorting device includes a sorting controller, a manipulator and a sorting line.
[0083] In a specific embodiment, when the drive motor receives a start pulse signal, the drive motor starts to operate, causing the conveyor belt to lift the packaged express parcels to be inspected and transmit them to various devices for identification of contraband.
[0084] The X-ray security inspection equipment of the scanning and detection device is used to scan the express parcels passing through the equipment to obtain X-ray images. The detection device uses a position sensor to determine the position information of the express parcel, starts the scanning controller, and records the serial number of the express parcel at the same time. At the same time, the express parcel scanning position signal and sequence signal are sent to the data collector to confirm the position and sequence information of the express parcel of prohibited items. Among them, the "position signal" refers to the position information of the current express parcel on the conveyor belt when it passes through the X-ray security inspection equipment. The express parcels are sorted in order when passing through the X-ray security inspection equipment, and the "sequence signal" refers to the order sorting information of the current express parcel passing through the X-ray security inspection equipment.
[0085] After receiving the express parcel location signal and sequence signal sent by the detection device, the data collector sends it to the data processing terminal for comparison with the identification label of the contraband express parcel. When the transmission controller receives the pulse signal from the data processing terminal, it controls the transmission device to start, and at the same time, the scanning controller and the sorting controller respectively control the operation of the scanning and detection device and the sorting device.
[0086] In this embodiment, the data processing terminal is a computer terminal, including a display screen, a computer server, and loaded image feature recognition software. The display screen is used to display the X-ray image and the express package detection effect diagram. The computer server is used to save the scanning results and perform detection operations. The image feature recognition software extracts features from the X-ray image to realize the target detection function.
[0087] The extracted features include: 1) improved local binary pattern features of the image; 2) directional gradient density features of the image.
[0088] The extraction method is: after programming the extraction algorithm on the computer, the input interface of the algorithm is set to the collected X-ray image source, and the algorithm processes the image to obtain the desired features.
[0089] Processing after extraction: After feature extraction, the extracted features are converted into vector form and input into the trained classifier. The classifier maps the features to the category to which they belong, and determines whether they are contraband. For example: the current X-ray image is an X-ray image of a battery. The feature extraction algorithm processes the image, converts its improved local binary pattern features and directional gradient density features into vector patterns, and inputs the vector into the classifier. The classifier determines whether the battery package represented by the vector is a contraband through the mapping of nonlinear functions.
[0090] When the robot receives the start signal from the sorting controller, it pushes the express parcel currently on the conveyor belt into the contraband sorting line.
[0091] In a specific embodiment, before the express parcel passes through the X-ray security inspection device, the first detection device obtains the location information and sequence information of the express parcel, and then the express parcel is conveyed to the X-ray security inspection device by the conveyor belt for security scanning and detection, and finally passes through the second detection device and combined with the sorting device to sort the express parcel for contraband. The first detection device and the second detection device are respectively installed at both ends of the electric conveyor belt and fixed by a scanning frame.
[0092] In a preferred embodiment, the robot and the sorting line are located on both sides of the conveyor belt. After receiving the signal from the sorting controller, the robot pushes the express parcel on the current conveyor belt into the sorting line.
[0093] Below, the specific embodiments of the present invention are described in further detail:
[0094] Step 1, the initial stage of security inspection, the transmission controller sends a pulse signal to the drive motor, and the electric transmission belt starts to operate.
[0095] Step 2: Put the packaged express parcel on the electric conveyor belt. When the express parcel passes the detection device, the sensor detects the express parcel passing by, starts the scanning controller, records the location information and sequence information of the express parcel, and sends it to the data collector. Then the express parcel enters the X-ray machine.
[0096] Step three, the X-ray machine emits X-rays to scan the express package, and the scan result is sent to the computer server. The X-ray scan inspection picture of the express package is displayed on the display screen through the imaging system.
[0097] Step 4: Input the X-ray image of the express package into the computer terminal, detect it through image feature recognition software, determine whether the express package contains prohibited items, write the X-ray image into the prohibited items category label, and display the detection result on the display screen.
[0098] It should be noted that the prohibited goods category labels are mainly divided into two categories: prohibited goods and non-prohibited goods. The features of these two types of sample data are extracted in advance to train the classifier. The output of the classifier is prohibited goods and non-prohibited goods, that is, transition information. Note: The expression of the prohibited goods category label here is incorrect, it would be better to write it as a category label.
[0099] The X-ray image of the express package is first converted into a vector form by a feature extraction algorithm, and then passed through a classifier. The classifier determines which category its feature composition is closer to, and then divides the image into that category, that is, labels it with that category (for example, the current package is a battery, and the battery passes through the security inspection machine to form an X-ray image. The feature extraction algorithm introduces the image and converts it into a vector form as the input of the classifier. The classifier determines that it belongs to the prohibited category through function mapping, labels it as a prohibited item, and displays the detection result.
[0100] In this part, the main steps of the detection algorithm are as follows:
[0101] 1) Establish an X-ray image database of express parcels, divide them into two categories: prohibited items and non-prohibited items, and set the prohibited items category as the target detection category;
[0102] 2) Divide the X-ray image into multiple connected regions, each of which is a 3*3 pixel block.
[0103] 3) Extract the features of each connected area. Specifically,
[0104] A. Use the following formula to calculate the local rotation-invariant and uniform-mode characteristic variables of the connected region:
[0105]
[0106] B. Set the transformation threshold of the feature variable parameters obtained in step A above to N times, and add N-dimensional features on the basis of the feature variable dimension. The formula is:
[0107] Among them, P is the number of pixels in the neighborhood, R is the neighborhood radius, and g p is the gray value of the neighborhood pixel, g c is the grayscale pixel value of the center point, S(x) is the sign function, and U(x) calculates the number of 0 / 1 transitions in the local binary pattern;
[0108] C. Select a gradient operator to calculate the directional gradient density feature histogram of the connected area:
[0109]
[0110] Where M(x,y) is the gradient magnitude of the pixel, θ(x,y) is the gradient direction of the pixel, and θ(x,y) is divided into several bins to obtain the directional gradient density feature histogram;
[0111] D. Combine the feature variables obtained in the above steps to construct a feature descriptor:
[0112] feature=concat(feature1,feature2,feature3,...,featureL)
[0113] 4) Construct a learning classifier, train the features represented by the feature descriptor, and learn the feature classification decision surface y=wx+b;
[0114] A. The dimension of the input vector is the dimension of the features represented by the feature descriptor;
[0115] B. Normalize the scale of feature variables:
[0116]
[0117] C. Design and use an enhanced kernel function based on Manhattan distance and Euclidean distance improvement, the formula is:
[0118]
[0119] The X-ray images of express packages are used as training data to train the learning classifier designed above to learn the feature classification decision surface;
[0120] 5) Access the real-time image data from the imaging system and perform cutting processing on the image based on the outline of the express package;
[0121] 6) Construct a feature descriptor for the cut express package image and put it into the trained learning model to obtain the prohibited item label information of the express package.
[0122] Step 5: The data collector sends the location information and sequence information of the express package to the computer server, and compares it with the contraband label of the X-ray scan image to determine the location information and sequence information of the express package containing contraband.
[0123] Step six: After the express parcel passes through the X-ray security inspection equipment, it enters the sorting unit. When the detection device detects that the express parcel has passed, it starts the scanning controller, records the location information and sequence information of the express parcel, and sends it to the computer server through the data collector.
[0124] Step 7: The computer server compares the location information and sequence information of the prohibited express parcel to determine whether the express parcel entering the sorting unit contains prohibited items. If the current express parcel contains prohibited items, a pulse signal is sent to the sorting controller to start the manipulator device.
[0125] Step 8: After receiving the start signal, the robot device pushes the express parcel containing prohibited items into the prohibited items sorting line.
[0126] Based on the teachings of the above embodiments, in the process of security inspection of large batches of express parcels, the existing technology requires a lot of manpower and material resources, and the cost remains high. The present invention mainly designs an intelligent sorting system for express parcels to reduce security inspection costs and improve the efficiency of security inspection of prohibited items.
[0127] The second embodiment of the present invention:
[0128] See also Figure 4 , an intelligent sorting method for express parcels based on X-ray images, suitable for a data processing terminal, comprising:
[0129] S10, receiving the location information, serial number information, and X-ray image of the express package transmitted by the scanning and detection device;
[0130] S11, detecting the express package according to a preset detection algorithm, and when it is determined that the express package contains contraband, writing the X-ray image into a contraband category label, displaying the detection result and giving an alarm;
[0131] S12: sending a sorting instruction carrying the location information and the serial number information to the sorting device, so that the express parcel is sorted out according to the sorting instruction.
[0132] Wherein, the conveying device includes a driving motor and a conveyor belt driven by the driving motor; the sorting device includes a sorting controller, a robot and a sorting line.
[0133] Among them, the scanning and detection device includes an X-ray security inspection device, and a first detection device and a second detection device respectively installed at both ends of the X-ray security inspection device along the direction of the conveyor belt; the first detector device includes a first position sensor, a first data collector and a first scanning controller, the second detector device includes a second position sensor, a second data collector and a second scanning controller, and the X-ray security inspection equipment includes a third data collector; the sorting device includes a sorting controller, a manipulator and a sorting line.
[0134] In a specific embodiment, when the drive motor receives a start pulse signal, the drive motor starts to operate, causing the conveyor belt to lift the packaged express parcels to be inspected and transmit them to various devices for identification of contraband.
[0135] The X-ray security inspection equipment of the scanning and detection device is used to scan the express parcels passing through the equipment to obtain X-ray images. The detection device uses a position sensor to determine the position information of the express parcel, starts the scanning controller, and records the serial number of the express parcel at the same time. At the same time, the express parcel scanning position signal and sequence signal are sent to the data collector to confirm the position and sequence information of the express parcel of prohibited items. Among them, the "position signal" refers to the position information of the current express parcel on the conveyor belt when it passes through the X-ray security inspection equipment. The express parcels are sorted in order when passing through the X-ray security inspection equipment, and the "sequence signal" refers to the order sorting information of the current express parcel passing through the X-ray security inspection equipment.
[0136] After receiving the express parcel location signal and sequence signal sent by the detection device, the data collector sends it to the data processing terminal for comparison with the identification label of the contraband express parcel. When the transmission controller receives the pulse signal from the data processing terminal, it controls the transmission device to start, and at the same time, the scanning controller and the sorting controller respectively control the operation of the scanning and detection device and the sorting device.
[0137] In this embodiment, the data processing terminal is a computer terminal, including a display screen, a computer server, and loaded image feature recognition software. The display screen is used to display the X-ray image and the express package detection effect diagram. The computer server is used to save the scanning results and perform detection operations. The image feature recognition software extracts features from the X-ray image to realize the target detection function.
[0138] The extracted features include: 1) improved local binary pattern features of the image; 2) directional gradient density features of the image.
[0139] The extraction method is: after programming the extraction algorithm on the computer, the input interface of the algorithm is set to the collected X-ray image source, and the algorithm processes the image to obtain the desired features.
[0140] Processing after extraction: After feature extraction, the extracted features are converted into vector form and input into the trained classifier. The classifier maps the features to the category to which they belong, and determines whether they are contraband. For example: the current X-ray image is an X-ray image of a battery. The feature extraction algorithm processes the image, converts its improved local binary pattern features and directional gradient density features into vector patterns, and inputs the vector into the classifier. The classifier determines whether the battery package represented by the vector is a contraband through the mapping of nonlinear functions.
[0141] When the robot receives the start signal from the sorting controller, it pushes the express parcel currently on the conveyor belt into the contraband sorting line.
[0142] In a specific embodiment, before the express parcel passes through the X-ray security inspection device, the first detection device obtains the location information and sequence information of the express parcel, and then the express parcel is conveyed to the X-ray security inspection device by the conveyor belt for security scanning and detection, and finally passes through the second detection device and combined with the sorting device to sort the express parcel for contraband. The first detection device and the second detection device are respectively installed at both ends of the electric conveyor belt and fixed by a scanning frame.
[0143] In a preferred embodiment, the robot and the sorting line are located on both sides of the conveyor belt. After receiving the signal from the sorting controller, the robot pushes the express parcel on the current conveyor belt into the sorting line.
[0144] Below, the specific embodiments of the present invention are described in further detail:
[0145] Step 1, the initial stage of security inspection, the transmission controller sends a pulse signal to the drive motor, and the electric transmission belt starts to operate.
[0146] Step 2: Put the packaged express parcel on the electric conveyor belt. When the express parcel passes the detection device, the sensor detects the express parcel passing by, starts the scanning controller, records the location information and sequence information of the express parcel, and sends it to the data collector. Then the express parcel enters the X-ray machine.
[0147] Step three, the X-ray machine emits X-rays to scan the express package, and the scan result is sent to the computer server. The X-ray scan inspection picture of the express package is displayed on the display screen through the imaging system.
[0148] Step 4: Input the X-ray image of the express package into the computer terminal, detect it through image feature recognition software, determine whether the express package contains prohibited items, write the X-ray image into the prohibited items category label, and display the detection result on the display screen.
[0149] It should be noted that the prohibited goods category labels are mainly divided into two categories: prohibited goods and non-prohibited goods. The features of these two types of sample data are extracted in advance to train the classifier. The output of the classifier is prohibited goods and non-prohibited goods, that is, transition information. Note: The expression of the prohibited goods category label here is incorrect, it would be better to write it as a category label.
[0150] The X-ray image of the express package is first converted into a vector form by a feature extraction algorithm, and then passed through a classifier. The classifier determines which category its feature composition is closer to, and then divides the image into that category, that is, labels it with that category (for example, the current package is a battery, and the battery passes through the security inspection machine to form an X-ray image. The feature extraction algorithm introduces the image and converts it into a vector form as the input of the classifier. The classifier determines that it belongs to the prohibited category through function mapping, labels it as a prohibited item, and displays the detection result.
[0151] In this part, the main steps of the detection algorithm are as follows:
[0152] 1) Establish an X-ray image database of express parcels, divide them into two categories: prohibited items and non-prohibited items, and set the prohibited items category as the target detection category;
[0153] 2) Divide the X-ray image into multiple connected regions, each of which is a 3*3 pixel block.
[0154] 3) Extract the features of each connected area. Specifically,
[0155] A. Use the following formula to calculate the local rotation-invariant and uniform-mode characteristic variables of the connected region:
[0156]
[0157] B. Set the transformation threshold of the feature variable parameters obtained in step A above to N times, and add N-dimensional features on the basis of the feature variable dimension. The formula is:
[0158] Among them, P is the number of pixels in the neighborhood, R is the neighborhood radius, and g p is the gray value of the neighborhood pixel, g c is the grayscale pixel value of the center point, S(x) is the sign function, and U(x) calculates the number of 0 / 1 transitions in the local binary pattern;
[0159] C. Select a gradient operator to calculate the directional gradient density feature histogram of the connected area:
[0160]
[0161] Where M(x,y) is the gradient magnitude of the pixel, θ(x,y) is the gradient direction of the pixel, and θ(x,y) is divided into several bins to obtain the directional gradient density feature histogram;
[0162] D. Combine the feature variables obtained in the above steps to construct a feature descriptor:
[0163] feature=concat(feature1,feature2,feature3,...,featureL)
[0164] 4) Construct a learning classifier, train the features represented by the feature descriptor, and learn the feature classification decision surface y=wx+b;
[0165] A. The dimension of the input vector is the dimension of the features represented by the feature descriptor;
[0166] B. Normalize the scale of feature variables:
[0167]
[0168] C. Design and use an enhanced kernel function based on Manhattan distance and Euclidean distance improvement, the formula is:
[0169]
[0170] The X-ray images of express packages are used as training data to train the learning classifier designed above to learn the feature classification decision surface;
[0171] 5) Access the real-time image data from the imaging system and perform cutting processing on the image based on the outline of the express package;
[0172] 6) Construct a feature descriptor for the cut express package image and put it into the trained learning model to obtain the prohibited item label information of the express package.
[0173] Step 5: The data collector sends the location information and sequence information of the express package to the computer server, and compares it with the contraband label of the X-ray scan image to determine the location information and sequence information of the express package containing contraband.
[0174] Step six: After the express parcel passes through the X-ray security inspection equipment, it enters the sorting unit. When the detection device detects that the express parcel has passed, it starts the scanning controller, records the location information and sequence information of the express parcel, and sends it to the computer server through the data collector.
[0175] Step 7: The computer server compares the location information and sequence information of the prohibited express parcel to determine whether the express parcel entering the sorting unit contains prohibited items. If the current express parcel contains prohibited items, a pulse signal is sent to the sorting controller to start the manipulator device.
[0176] Step 8: After receiving the start signal, the robot device pushes the express parcel containing prohibited items into the prohibited items sorting line.
[0177] Based on the teachings of the above embodiments, the present invention can improve the inspection efficiency and accuracy of express packages, and accurately identify and sort prohibited items.
[0178] In a third aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the storage medium includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the above-mentioned X-ray image-based intelligent sorting method for express parcels.
[0179] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.
[0180] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.
Claims
1. An intelligent express parcel sorting system based on X-ray images, characterized in that: It includes a data processing terminal, a conveying device, a scanning and detecting device, and a sorting device; The express parcel is conveyed by the conveying device, passes through the scanning and detecting device, and then arrives at the sorting device; The scanning and detection device is used to collect the location information, serial number information, and X-ray image of the express package and transmit them to the data processing terminal; The data processing terminal is used to detect the express package according to a preset detection algorithm, and when it is determined that the express package contains contraband, write the X-ray image into a contraband category label, display the detection result and issue an alarm; and, sending a sorting instruction carrying the location information and the serial number information to the sorting device; The sorting device is used to sort out the express parcel according to the sorting instruction; Wherein, the preset detection algorithm includes: 1) Establish an X-ray image database of express parcels, divide them into two categories: prohibited items and non-prohibited items, and set the prohibited items category as the target detection category; 2) Divide the X-ray image into multiple connected regions, each of which is a 3*3 pixel block; 3) Extract the features of each connected area. Specifically, A. Use the following formula to calculate the local rotation-invariant and uniform-mode characteristic variables of the connected region: B. Set the transformation threshold of the feature variable parameters obtained in step A above to N times, and add N-dimensional features on the basis of the feature variable dimension. The formula is: Among them, P is the number of pixels in the neighborhood, R is the neighborhood radius, and g p is the gray value of the neighborhood pixel, g c is the grayscale pixel value of the center point, S(x) is the sign function, and U(x) calculates the number of 0 / 1 transitions in the local binary pattern; C. Select a gradient operator to calculate the directional gradient density feature histogram of the connected area: Where M(x,y) is the gradient magnitude of the pixel, θ(x,y) is the gradient direction of the pixel, and θ(x,y) is divided into several bins to obtain the directional gradient density feature histogram; D. Combine the feature variables obtained in the above steps to construct a feature descriptor: feature=concat(feature1,feature2,feature3,...,featureL); 4) Construct a learning classifier, train the features represented by the feature descriptor, and learn the feature classification decision surface y=wx+b; A. The dimension of the input vector is the dimension of the features represented by the feature descriptor; B. Normalize the scale of feature variables: C. Design and use an enhanced kernel function based on Manhattan distance and Euclidean distance improvement, the formula is: The X-ray images of express packages are used as training data to train the learning classifier designed above to learn the feature classification decision surface; 5) Access the real-time image data from the imaging system and perform cutting processing on the image based on the outline of the express package; 6) Construct a feature descriptor for the cut express package image and put it into the trained learning model to obtain the prohibited item label information of the express package.
2. The intelligent express parcel sorting system based on X-ray images as claimed in claim 1, characterized in that: The conveying device includes a driving motor and a conveyor belt driven by the driving motor; the sorting device includes a sorting controller, a robot and a sorting line.
3. The intelligent express parcel sorting system based on X-ray images as claimed in claim 1, characterized in that: The scanning and detection device includes an X-ray security inspection device, and a first detection device and a second detection device respectively installed at both ends of the X-ray security inspection device along the conveyor belt direction; the first detection device includes a first position sensor, a first data collector and a first scanning controller, the second detection device includes a second position sensor, a second data collector and a second scanning controller, and the X-ray security inspection equipment includes a third data collector; the sorting device includes a sorting controller, a manipulator and a sorting line.
4. An intelligent sorting method for express parcels based on X-ray images, suitable for data processing terminals, characterized in that: include: Receiving the location information, serial number information, and X-ray image of the express package transmitted by the scanning and detection device; The express package is inspected according to a preset inspection algorithm. When it is determined that the express package contains contraband, the X-ray image is written into a contraband category label, the inspection result is displayed, and an alarm is issued; Sending a sorting instruction carrying the location information and the serial number information to a sorting device, so that the sorting device sorts out the express parcel according to the sorting instruction; Wherein, the preset detection algorithm includes: 1) Establish an X-ray image database of express parcels, divide them into two categories: prohibited items and non-prohibited items, and set the prohibited items category as the target detection category; 2) Divide the X-ray image into multiple connected regions, each of which is a 3*3 pixel block; 3) Extract the features of each connected area. Specifically, A. Use the following formula to calculate the local rotation-invariant and uniform-mode characteristic variables of the connected region: B. Set the transformation threshold of the feature variable parameters obtained in step A above to N times, and add N-dimensional features on the basis of the feature variable dimension. The formula is: Among them, P is the number of pixels in the neighborhood, R is the neighborhood radius, and g p is the gray value of the neighborhood pixel, g c is the grayscale pixel value of the center point, S(x) is the sign function, and U(x) calculates the number of 0 / 1 transitions in the local binary pattern; C. Select a gradient operator to calculate the directional gradient density feature histogram of the connected area: Where M(x,y) is the gradient magnitude of the pixel, θ(x,y) is the gradient direction of the pixel, and θ(x,y) is divided into several bins to obtain the directional gradient density feature histogram; D. Combine the feature variables obtained in the above steps to construct a feature descriptor: feature=concat(feature1,feature2,feature3,...,featureL); 4) Construct a learning classifier, train the features represented by the feature descriptor, and learn the feature classification decision surface y=wx+b; A. The dimension of the input vector is the dimension of the features represented by the feature descriptor; B. Normalize the scale of feature variables: C. Design and use an enhanced kernel function based on Manhattan distance and Euclidean distance improvement, the formula is: The X-ray images of express packages are used as training data to train the learning classifier designed above to learn the feature classification decision surface; 5) Access the real-time image data from the imaging system and perform cutting processing on the image based on the outline of the express package; 6) Construct a feature descriptor for the cut express package image and put it into the trained learning model to obtain the prohibited item label information of the express package.
5. The method for intelligent sorting of express parcels based on X-ray images as claimed in claim 4, characterized in that: The scanning and detection device includes an X-ray security inspection device, and a first detection device and a second detection device respectively installed at both ends of the X-ray security inspection device along the conveyor belt direction; the first detection device includes a first position sensor, a first data collector and a first scanning controller, the second detection device includes a second position sensor, a second data collector and a second scanning controller, and the X-ray security inspection equipment includes a third data collector; the sorting device includes a sorting controller, a manipulator and a sorting line.
6. A computer-readable storage medium, characterized in that: The storage medium includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the express parcel intelligent sorting method based on X-ray images as described in any one of claims 4 to 5.
Citation Information
Patent Citations
High-speed X-ray package sorting method
CN107999413A
Scale adaptive target tracking method based on fast compressive tracking algorithm
CN108062557A
Contraband intelligence letter sorting system
CN206184803U
Express mail is tracked and automated sorting system based on raster scan
CN206325847U
Intelligent express parcel sorting system based on X-ray image
CN211275544U
