A parcel image segmentation method, device, system and image processing equipment

Abstract

The embodiment of the present application provides a parcel image segmentation method, device, system and image processing equipment, the method is applied to the image processing equipment in the security inspection system, the security inspection system further includes a security inspection equipment, the security inspection equipment has a detector and a detection unit, the detection unit includes multiple groups of detectors, the detection range of the multiple groups of detectors covers the space of the preset height perpendicular to the conveying belt transmission direction of the security inspection equipment, and the method comprises the following steps: obtaining parcel data detected by the detector in the operation process of the security inspection equipment and a detection result of the detection unit; in the case that the detection result indicates that there is no parcel on the conveying belt, determining parcel data in the parcel data based on the difference between the parcel data and reference data; and based on the parcel data, the parcel data is segmented to obtain a parcel image. The embodiment of the present application can improve the accuracy of parcel image segmentation and reduce the loss of parcel data.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method, apparatus, system and image processing device for segmenting wrapped images. Background Technology

[0002] X-ray security inspection equipment, which uses X-rays to visualize packages, is widely used in various locations. For example, in airports and subways, security equipment can inspect packages carried by people entering the premises and obtain images of the packages.

[0003] During the continuous passage of bags through X-ray security inspection equipment, a bag data stream is obtained, and a package image can be generated based on this data stream. This data stream contains not only actual package data but also an "air value." The air value refers to the data detected by the detector when there are no packages in the X-ray inspection channel. To segment the actual package data, a pair of infrared light barriers are currently installed on both sides of the conveyor belt. Each triggering of the infrared light barrier yields a segmented package image.

[0004] However, due to the limited coverage of infrared light barriers, for some irregularly shaped packages, such as chairs, tables, and toys, the infrared light barriers will be triggered multiple times during the entire package passage process. This results in a package being segmented into multiple images, which will lose some effective data and lead to low package image segmentation accuracy. Summary of the Invention

[0005] The purpose of this application is to provide a method, apparatus, system, image processing device, and storage medium for package image segmentation, so as to improve the accuracy of package image segmentation. The specific technical solution is as follows:

[0006] In a first aspect, embodiments of this application provide a package image segmentation method applied to an image processing device in a security inspection system. The security inspection system further includes security inspection equipment, which has detectors and a detection unit. The detection unit includes multiple sets of detectors, the detection range of which covers a space at a preset height perpendicular to the conveyor belt transport direction of the security inspection equipment. The method includes:

[0007] The system acquires the bag data detected by the detector and the detection results of the detection unit during the operation of the security inspection equipment, wherein the detection unit is used to detect whether there are bags on the conveyor belt of the security inspection equipment.

[0008] If the detection result indicates that there are no packages on the conveyor belt, the package data in the package passing data is determined based on the difference between the package passing data and the reference data. The reference data is the data detected by the detector when there are no packages on the conveyor belt, which is obtained in advance. The package data refers to the detection data in the package passing data that is judged to belong to a package.

[0009] Based on the package data, the package data is segmented to obtain a package image.

[0010] Optionally, the detector is a linear array detector; the step of determining the package data in the over-packet data based on the difference between the over-packet data and the reference data includes:

[0011] Calculate the difference between each column of data and the reference data, and determine whether each component of the difference is not less than a preset threshold, wherein the preset threshold is determined based on the value of the reference data;

[0012] If there is a target component that is not less than the preset threshold, determine whether the number of the target component meets the preset number condition, wherein the preset number condition is determined based on the minimum number of pixels in a column of pixels when the actual image is encapsulated in the detector;

[0013] If the conditions are met, then the data in this column is determined to be package data.

[0014] Optionally, the step of determining whether the quantity of the target component meets the preset quantity condition includes:

[0015] Determine whether the total number of target components in the difference reaches a first preset number; and determine whether the total number of consecutive components in the target components reaches a second preset number, wherein the consecutive components are components whose corresponding pixel points are consecutive pixels in the packet data;

[0016] If the total number of the target components reaches a first preset number, and the total number of the consecutive components reaches a second preset number, then the data in that column is determined to be package data.

[0017] Optionally, the method further includes:

[0018] If there are no packages on the conveyor belt of the security inspection equipment, a series of data detected by the linear array detector is acquired as the reference data; or,

[0019] If there are no packages on the conveyor belt of the security inspection equipment, multiple columns of data detected by the linear array detector are acquired, and the average value of the multiple columns of data is calculated as the reference data.

[0020] Optionally, the method further includes:

[0021] If the detection result of at least one of the multiple sets of detectors indicates that there is a package on the conveyor belt, the package data is determined to be package data.

[0022] Optionally, the multiple sets of detectors include at least a first set of detectors and a second set of detectors. The first set of detectors includes a first transmitter, a first receiver, and a second receiver. The second set of detectors includes a second transmitter, a third receiver, and a fourth receiver.

[0023] The first transmitter and the second transmitter are respectively disposed on both sides along the width direction of the conveyor belt, and the first receiver and the second receiver are located on the opposite side of the first transmitter. The first receiver and the second receiver are disposed at different heights perpendicular to the conveyor belt transmission direction. The first receiver and the second receiver are used to receive the detection signal emitted by the first transmitter.

[0024] The third receiver and the fourth receiver are located on the opposite side of the second transmitter. The third receiver and the fourth receiver are set at different heights perpendicular to the conveyor belt transmission direction, and are respectively used to receive the detection signal emitted by the second transmitter.

[0025] The detection range of the first group of detectors does not completely overlap with the detection range of the second group of detectors.

[0026] Optionally, the step of segmenting the package data to obtain a package image based on the package data includes:

[0027] The consecutive parcel data in the parcel passing data are segmented into the same parcel image.

[0028] Secondly, embodiments of this application provide a package image segmentation device, which is configured in an image processing device within a security inspection system. The security inspection system further includes security inspection equipment, which has detectors and a detection unit. The detection unit includes multiple sets of detectors, the detection range of which covers a predetermined height perpendicular to the conveyor belt transport direction of the security inspection equipment. The device includes:

[0029] The package data and detection result acquisition module is used to acquire the package data detected by the detector and the detection result of the detection unit during the operation of the security inspection equipment, wherein the detection unit is used to detect whether there is a package on the conveyor belt of the security inspection equipment;

[0030] The first determining module is used to determine package data in the package passing data based on the difference between the package passing data and the reference data when the detection result indicates that there are no packages on the conveyor belt. The reference data is the data detected by the detector when there are no packages on the conveyor belt, which is obtained in advance. The package data refers to the detection data in the package passing data that is judged to belong to a package.

[0031] The package image segmentation module is used to segment the package data based on the package data to obtain a package image.

[0032] Optionally, the detector is a linear array detector; the first determining module includes:

[0033] The preset threshold judgment submodule is used to calculate the difference between each column of over-packed data and the reference data, and to determine whether each component of the difference is not less than a preset threshold, wherein the preset threshold is determined based on the value of the reference data;

[0034] The preset quantity condition judgment submodule is used to determine whether the quantity of the target component satisfies the preset quantity condition if there is a target component that is not less than the preset threshold. The preset quantity condition is determined based on the minimum number of pixels in a column of pixels when the actual image is encapsulated in the detector.

[0035] The first determination submodule is used to determine if the data in the column is package data if certain conditions are met.

[0036] Optionally, the preset quantity condition judgment submodule includes:

[0037] A preset quantity judgment unit is used to determine whether the total number of target components in the difference reaches a first preset quantity; and to determine whether the total number of continuous components in the target components reaches a second preset quantity, wherein the continuous component is a component whose corresponding pixel point in the packet data is a continuous pixel point;

[0038] The package data determination unit is used to determine that the column of package data is package data if the total number of the target components reaches a first preset number and the total number of the consecutive components reaches a second preset number.

[0039] Optionally, the device further includes:

[0040] The reference data acquisition module is used to acquire a series of data detected by the linear array detector as the reference data when there are no packages on the conveyor belt of the security inspection equipment; or,

[0041] This is used to acquire multiple columns of data detected by the linear array detector when there are no packages on the conveyor belt of the security inspection equipment, and to calculate the average value of the multiple columns of data as the reference data.

[0042] Optionally, the device further includes:

[0043] The second determining module is used to determine that the package passing data is package data when the detection results of at least one of the multiple sets of detectors indicate that there is a package on the conveyor belt.

[0044] Optionally, the multiple sets of detectors include at least a first set of detectors and a second set of detectors. The first set of detectors includes a first transmitter, a first receiver, and a second receiver. The second set of detectors includes a second transmitter, a third receiver, and a fourth receiver.

[0045] The first transmitter and the second transmitter are respectively disposed on both sides along the width direction of the conveyor belt, and the first receiver and the second receiver are located on the opposite side of the first transmitter. The first receiver and the second receiver are disposed at different heights perpendicular to the conveyor belt transmission direction. The first receiver and the second receiver are used to receive the detection signal emitted by the first transmitter.

[0046] The third receiver and the fourth receiver are located on the opposite side of the second transmitter. The third receiver and the fourth receiver are set at different heights perpendicular to the conveyor belt transmission direction, and are respectively used to receive the detection signal emitted by the second transmitter.

[0047] The detection range of the first group of detectors does not completely overlap with the detection range of the second group of detectors.

[0048] Optionally, the package image segmentation module includes:

[0049] The package image segmentation submodule is used to segment consecutive package data in the package data into the same package image.

[0050] Thirdly, embodiments of this application provide a security inspection system, which includes an image processing device and a security inspection device. The security inspection device has a detector and a detection unit. The detection unit includes multiple sets of detectors, and the detection range of the multiple sets of detectors covers a space at a preset height perpendicular to the conveyor belt transport direction of the security inspection device, wherein:

[0051] The detection unit is used to detect whether there are packages on the conveyor belt of the security inspection equipment and obtain the detection result;

[0052] The detector is used to detect packages on the conveyor belt of the security inspection equipment during the operation of the security inspection equipment, and to obtain package data.

[0053] The image processing device is used to acquire the package passing data and the detection result; when the detection result indicates that there are no packages on the conveyor belt, it determines the package data in the package passing data based on the difference between the package passing data and the reference data; based on the package data, it segments the package passing data to obtain a package image, wherein the reference data is the data detected by the detector when there are no packages on the conveyor belt, and the package data refers to the detection data in the package passing data that is judged to belong to a package.

[0054] Optionally, the detector is a linear array detector;

[0055] The image processing device is specifically used to calculate the difference between each column of overpacking data and reference data, determine whether each component in the difference is not less than a preset threshold, and if there is a target component not less than the preset threshold, determine whether the number of the target component meets a preset quantity condition. If it does, determine that the column of overpacking data is package data. The preset threshold is determined based on the value of the reference data, and the preset quantity condition is determined based on the minimum number of pixels in a column of pixels when the actual package is imaged in the detector.

[0056] Optionally, the image processing device is specifically used to determine whether the total number of target components in the difference reaches a first preset number; and to determine whether the total number of consecutive components in the target components reaches a second preset number. If the total number of target components reaches the first preset number and the total number of consecutive components reaches the second preset number, the column of overpacking data is determined to be package data, wherein the consecutive components are components in the overpacking data whose corresponding pixel points are consecutive pixels.

[0057] Optionally, the image processing device is further configured to acquire a column of data detected by the linear array detector as the reference data when there are no packages on the conveyor belt of the security inspection equipment; or, it can also be configured to acquire multiple columns of data detected by the linear array detector and calculate the average value of the multiple columns of data as the reference data when there are no packages on the conveyor belt of the security inspection equipment.

[0058] Optionally, the image processing device is specifically used to determine that the package passing data is package data when the detection results of at least one of the multiple sets of detectors indicate that there is a package on the conveyor belt.

[0059] Optionally, the multiple sets of detectors include at least a first set of detectors and a second set of detectors. The first set of detectors includes a first transmitter, a first receiver, and a second receiver. The second set of detectors includes a second transmitter, a third receiver, and a fourth receiver.

[0060] The first transmitter and the second transmitter are respectively disposed on both sides along the width direction of the conveyor belt, and the first receiver and the second receiver are located on the opposite side of the first transmitter. The first receiver and the second receiver are disposed at different heights perpendicular to the conveyor belt transmission direction. The first receiver and the second receiver are used to receive the detection signal emitted by the first transmitter.

[0061] The third receiver and the fourth receiver are located on the opposite side of the second transmitter. The third receiver and the fourth receiver are set at different heights perpendicular to the conveyor belt transmission direction, and are respectively used to receive the detection signal emitted by the second transmitter.

[0062] The detection range of the first group of detectors does not completely overlap with the detection range of the second group of detectors.

[0063] Optionally, the image processing device is specifically used to segment the continuous package data in the package data into the same package image.

[0064] Fourthly, embodiments of this application provide an image processing device applied to a security inspection system. The security inspection system further includes security inspection equipment, which has detectors and a detection unit. The detection unit includes multiple sets of detectors, the detection range of which covers a preset height perpendicular to the conveyor belt transport direction of the security inspection equipment. The image processing device includes:

[0065] Memory, used to store computer programs;

[0066] When a processor executes a program stored in memory, it implements any of the methods described in the first aspect above.

[0067] Fifthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the methods described in the first aspect above.

[0068] Beneficial effects of the embodiments in this application:

[0069] The solution provided in this application embodiment is applied to an image processing device in a security inspection system. The security inspection system also includes security inspection equipment, which has detectors and detection units. The detection unit includes multiple sets of detectors, and the detection range of the multiple sets of detectors covers a space at a preset height perpendicular to the conveyor belt transmission direction of the security inspection equipment. The image processing device can acquire the bag-passing data detected by the detectors and the detection results of the detection units during the operation of the security inspection equipment. The detection unit is used to detect whether there are packages on the conveyor belt of the security inspection equipment. When the detection result indicates that there are no packages on the conveyor belt, the package data in the bag-passing data is determined based on the difference between the bag-passing data and the reference data. The reference data is the data detected by the detectors when there are no packages on the conveyor belt, which is obtained in advance. The package data refers to the detection data in the bag-passing data that is judged to belong to packages. Based on the package data, the bag-passing data is segmented to obtain a package image. Since the image processing device can further determine the package data in the package-passing data based on the difference between the package-passing data detected by the detector and the reference data when the detection unit detects no packages on the conveyor belt, it can segment the package-passing data to obtain a package image. This allows for accurate identification of package data in the package-passing data, even for irregularly shaped packages or other packages that might cause inaccurate detection results from the detection unit. It can accurately segment package data corresponding to a single package into a single package image, improving the accuracy of package image segmentation and reducing the loss of package data. Of course, implementing any product or method of this application does not necessarily require achieving all of the advantages described above simultaneously. Attached Figure Description

[0070] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0071] Figure 1 A flowchart illustrating a package image segmentation method provided in an embodiment of this application;

[0072] Figure 2 This is a schematic diagram of a package segmentation image provided in an embodiment of this application;

[0073] Figure 3 This is a schematic diagram of the structure of a security inspection device provided in an embodiment of this application;

[0074] Figure 4 for Figure 1 A specific flowchart of step S102 in the illustrated embodiment;

[0075] Figure 5 for Figure 4 A specific flowchart of step S402 in the illustrated embodiment;

[0076] Figure 6 This is a schematic diagram of the structure of the detection unit provided in an embodiment of this application;

[0077] Figure 7 This is another schematic diagram of a package segmentation image provided in an embodiment of this application;

[0078] Figure 8 This is a specific flowchart of the package image segmentation method provided in the embodiments of this application;

[0079] Figure 9 This is another specific flowchart of the package image segmentation method provided in the embodiments of this application;

[0080] Figure 10 This is a schematic diagram of the structure of a package image segmentation device provided in an embodiment of this application;

[0081] Figure 11 This is a schematic diagram of the structure of a security inspection system provided in an embodiment of this application;

[0082] Figure 12 This is a schematic diagram of the structure of an image processing device provided in an embodiment of this application. Detailed Implementation

[0083] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0084] To improve the accuracy of package image segmentation, embodiments of this application provide a package image segmentation method, apparatus, system, image processing device, computer-readable storage medium, and computer program product. The package image segmentation method provided in this application embodiment is described below.

[0085] The package image segmentation method provided in this application can be applied to image processing equipment in a security inspection system that requires package image segmentation. The security inspection system also includes security inspection equipment, which has detectors and detection units. The detection unit includes multiple sets of detectors, and the detection range of these multiple sets of detectors covers a preset height of space perpendicular to the conveyor belt transport direction of the security inspection equipment.

[0086] like Figure 1As shown, a package image segmentation method is applied to an image processing device in a security inspection system. The method includes:

[0087] S101, acquire the bag-passing data detected by the detector and the detection results of the detection unit during the operation of the security inspection equipment;

[0088] The detection unit is used to detect whether there are packages on the conveyor belt of the security inspection equipment.

[0089] S102, if the detection result indicates that there are no packages on the conveyor belt, determine the package data in the package data based on the difference between the package passing data and the reference data;

[0090] The reference data refers to the data detected by the detector when there are no packages on the conveyor belt, which is obtained in advance. The package data refers to the detection data in the package passing data that is determined to belong to a package.

[0091] S103, based on the package data, the package data is segmented to obtain a package image.

[0092] As can be seen, in the solution provided in this application embodiment, the image processing device is applied to the security inspection system. The security inspection system also includes security inspection equipment, which has detectors and detection units. The detection unit includes multiple sets of detectors. The detection range of the multiple sets of detectors covers a space at a preset height perpendicular to the conveyor belt transmission direction of the security inspection equipment. The image processing device can acquire the baggage data detected by the detectors and the detection results of the detection units during the operation of the security inspection equipment. The detection unit is used to detect whether there are packages on the conveyor belt of the security inspection equipment. When the detection result indicates that there are no packages on the conveyor belt, the package data in the baggage data is determined based on the difference between the baggage data and the reference data. The reference data is the data detected by the detectors when there are no packages on the conveyor belt, which is obtained in advance. The package data refers to the detection data in the baggage data that is judged to belong to packages. Based on the package data, the baggage data is segmented to obtain a package image. Since the detection unit detects no packages on the conveyor belt, the image processing device can further determine the package data in the package passing data based on the difference between the package passing data detected by the detector and the reference data. This allows the package passing data to be segmented to obtain a package image. In this way, even for irregularly shaped packages or other packages that may cause inaccurate detection results from the detection unit, the package data in the package passing data can be accurately determined. This enables the package data corresponding to a single package to be accurately segmented into a single package image, improving the accuracy of package image segmentation and reducing the loss of package data.

[0093] In public places such as subways, train stations, and large event venues, security screening equipment can detect packages carried by people entering the premises and obtain images of the packages. Detection units are typically installed on both sides of the conveyor belt across the width of the security screening equipment. Each detection unit can be a pair of infrared light barriers, consisting of a transmitter and a receiver. The transmitter emits infrared light to the receiver, which receives the emitted infrared light. When a package moves along the conveyor belt to the position of the infrared light barrier, the package blocks the infrared light, and the receiver of the infrared light barrier cannot receive the infrared light, thus triggering the infrared light barrier. Therefore, each triggering of the infrared light barrier indicates the detection of a package, and the image processing equipment can segment and obtain an image of the package when the infrared light barrier is triggered.

[0094] However, due to the limited detection range of infrared light barriers, for some irregularly shaped packages, the infrared light barrier may be triggered multiple times, causing a package to be segmented into multiple images, resulting in data loss and low package image segmentation accuracy. For example, if the package is a height-adjustable chair with a rectangular backrest that is hollow in the middle, and the seat and legs are connected by a height-adjustable rod, and the chair is placed on a conveyor belt, when it passes through an infrared light barrier, because the hollow backrest does not block infrared light, and the thin height-adjustable rod also does not block infrared light, the infrared light barrier will be triggered multiple times as the chair passes through. Figure 2 As shown, the chair image is divided into three package images, namely package images 201, 202, and 203.

[0095] To improve the accuracy of package image segmentation, the detection unit provided in this application embodiment includes multiple sets of detectors. These detectors can be strategically positioned on both sides of the conveyor belt width direction based on the height of the package being detected, so that the detection range of the multiple detectors can cover a preset height of space perpendicular to the conveyor belt's transport direction, thereby increasing the coverable detection range. In step S101 of the package image segmentation method provided in this application embodiment, the image processing device can acquire package data detected by the detectors and the detection results of the detection unit during the operation of the security inspection equipment.

[0096] During the baggage screening process at security checkpoints, detectors scan the area within the screening channel to obtain baggage data. This data can include package details and air density. Air density refers to the data detected by the detector when there are no packages in the screening channel. For example, if two packages are placed at a certain distance on a conveyor belt and move along with it, the data detected by the detector will include both package details and the air density between them.

[0097] During the operation of the security inspection equipment, the detection unit can detect whether there are packages on the conveyor belt, and the image processing equipment can acquire the detection results. If there are no packages on the conveyor belt, the detection unit can indicate that no packages are present; if there are packages on the conveyor belt, the detection unit can also indicate that packages are present. For example, the detection unit includes multiple sets of infrared light barriers. If there are no packages on the conveyor belt, none of the infrared light barriers will be triggered. If there are packages on the conveyor belt, at least one of the infrared light barriers will be triggered, indicating the presence of a package.

[0098] like Figure 3 As shown, a package 301 is on the conveyor belt of the security inspection equipment. Infrared light barriers 302 are installed on both sides of the conveyor belt in the width direction. The conveyor belt runs in the direction indicated by arrow 303. When the package passes through the infrared light barrier 302, it blocks the infrared light, triggering the infrared light barrier and indicating the presence of a package on the conveyor belt. The image processing equipment can then obtain the detection result from the detection unit. Furthermore, the detector can detect package data during the operation of the security inspection equipment. When package 301 passes through the detector's detection range, the package data includes the package data of package 301 and the air value. The image processing equipment can then obtain the package data detected by the detector.

[0099] After acquiring the detection results from the detection unit, the image processing device can determine whether there are packages on the conveyor belt based on the detection results. However, due to the presence of some extremely irregularly shaped packages on the conveyor belt, the detection results of multiple detectors are not accurate enough. In cases where packages are present on the conveyor belt, the device may indicate that no packages are present. Therefore, in step S102, even if the detection results indicate that no packages are present on the conveyor belt, the image processing device can determine the package data in the package data based on the difference between the package data and the reference data.

[0100] The reference data consists of data detected by the detector when there are no packages on the conveyor belt, while the package data refers to the detection data identified as packages within the package-passing data. Typically, the structure of a package differs from that of the conveyor belt. Therefore, the data detected by the detector when there are packages on the conveyor belt differs from the data when there are no packages. Based on the difference between the package-passing data and the reference data, it is possible to determine which data in the package-passing data represents actual packages and which represents empty packages.

[0101] For example, the package on the conveyor belt contains toys with extremely irregular shapes. The detection results indicate that some areas of the toy do not contain packages. When the toy moves into the detection range of the detector as the conveyor belt runs, the data detected by the detector includes toy data. The data detected by the detector differs from the reference data. Therefore, the image processing device can determine the package data in the data detected by ...

[0102] After the image processing device determines the package data, in step S103, the package data can be segmented to obtain a package image based on the package data.

[0103] Package image segmentation refers to the process by which image processing equipment can remove air values ​​before and after the package data from the acquired package data during the package passing process, thereby segmenting the package data into a single package image.

[0104] In one implementation, after the image processing device determines the package data, it can mark the package data. Then, based on the marking of the package data, the package data can be segmented to obtain a package image.

[0105] For example, if the detector is a linear array detector, the packet overpass data detected by the linear array detector is array data. The image processing device can identify the packet data in each column of packet overpass data, mark the packet data in each column of packet overpass data, and then segment the packet overpass data based on the markings of the packet data in each column of packet overpass data to obtain a packet image.

[0106] In another implementation, after the image processing device determines the package data, it can remove the air value from the package data and save the package data in the package data. Then, based on the saved package data, the package data can be segmented to obtain the package image.

[0107] For example, if the package placed on the conveyor belt of the security inspection equipment is a chair, after the image processing equipment determines the chair data, it can save the chair data. Based on the saved chair data, the data of the package detected by the detector can be segmented to obtain a complete image of the chair.

[0108] In this embodiment, when the detection unit detects that there are no packages on the conveyor belt, the image processing device can further determine the package data in the package data based on the difference between the package data detected by the detector and the reference data. This allows the package data to be segmented to obtain a package image. In this way, even for irregularly shaped packages or other packages that may cause inaccurate detection results from the detection unit, the package data in the package data can be accurately determined. This enables the package data corresponding to a package to be accurately segmented into a package image, improving the accuracy of package image segmentation and reducing the loss of package data.

[0109] As one embodiment of this application, the detector described above can be a linear array detector. As a type of detector, the linear array detector can achieve linear scanning imaging through the relative movement between the package and the detector. Therefore, the package-passing data detected by the linear array detector can be array data, and both the package-passing data and the reference data are composed of multiple vectors. Each column of data is a vector, and each column represents data obtained by the linear array detector detecting a region of a preset width on the conveyor belt perpendicular to the conveyor belt's running direction. Each element in the vector corresponds to the pixel value of the corresponding pixel in the package image.

[0110] For the case where the detector is a linear array detector, such as Figure 4 As shown, the step of determining the package data in the package transfer data based on the difference between the transfer data and the reference data may include:

[0111] S401, calculate the difference between each column of data and the reference data, and determine whether each component of the difference is not less than a preset threshold; if there is a target component that is not less than the preset threshold, proceed to step S402; if there is no target component that is not less than the preset threshold, proceed to step S404.

[0112] Since the package data includes both package data and air values, with the air values ​​serving as reference data, the larger the difference between any column of package data and the reference data, the greater the difference between the object structure in the conveyor belt area corresponding to that column of package data and the conveyor belt structure represented by the reference data. Therefore, the greater the likelihood that that column of package data represents package data. Thus, the image processing device can calculate the difference between each column of package data and the reference data, that is, calculate the difference between the value at each corresponding position of each component in each column of package data and the value at each corresponding position of each component in the reference data, and then determine whether each component in the difference is not less than a preset threshold.

[0113] The preset threshold is determined based on the value of the reference data. For example, it can be 5% or 10% of the reference data value, and can be adjusted according to the actual situation; no specific limit is set here. The number of pixels in each column detected by the detector can be determined according to the actual situation; no specific limit is set here.

[0114] For example, if the detector detects N pixels in each column, the packet data is a vector. The reference data is a vector. The preset threshold is 10% of the reference data value. The image processing device calculates the difference between the packet data and the reference data, i.e., it calculates the vector. with vector The difference between the components can be used to obtain a vector. The preset threshold is 10% * vector. Then, we can determine whether each component of the difference is not less than a preset threshold, that is, whether s1 is not less than 10%*p1, s2 is not less than 10%*p2, s3 is not less than 10%*p3…s k Is it not less than 10%? k …s N Is it not less than 10%? N .

[0115] If there is a target component that is not less than the preset threshold, proceed to step S402; if there is no target component that is not less than the preset threshold, determine that the data in that column is an empty value.

[0116] S402, determine whether the quantity of the target component meets the preset quantity condition; if it does, proceed to step S403; if it does not, proceed to step S405.

[0117] S403, confirm that the data in this column is package data;

[0118] S404, indicating that the data in this column is an empty value;

[0119] S405, confirm that the data in this column is not parcel data.

[0120] If any component in the difference has a value not less than a preset threshold, that component can be considered a target component. The presence of a target component indicates that the data in that column might be packaged data. However, the packaged data may contain noise or errors caused by environmental factors. The number of pixels caused by noise or other factors is generally single or a few. To reduce the impact of noise or errors, it can be further determined whether the number of target components meets a preset quantity condition. If it does, the data in that column can be determined to be packaged data; otherwise, it is determined that the data in that column is not packaged data.

[0121] The preset quantity condition is determined based on the minimum number of pixels corresponding to a single pixel in a column of pixels when the actual package is imaged in the detector. Specifically, due to structural and shape limitations, the actual package occupies a certain area on the conveyor belt. Even a very fine structure cannot occupy only a few pixels when detected and imaged by the detector. Therefore, the minimum number of pixels corresponding to the actual package in a column of pixels will definitely differ from the number of pixels formed by noise and other factors. Thus, the preset quantity condition can be determined based on the minimum number of pixels corresponding to the actual package in a column of pixels when imaged in the detector.

[0122] For example, if the detector detects 100 pixels in each column, the difference between the packet data and the reference data is calculated as a vector. Among them, the difference Each component in the equation has a component that is not less than a preset threshold, and these components are s1 to s2. 30 s 40 To s 60 s 70 To s 90 Then, the image processing device can determine whether the number of target components meets the preset quantity condition, that is, determine s1 to s2. 30 s 40 To s 60 s 70 To s 90 If the quantity meets the preset quantity condition, it can be determined that the data in this column is package data.

[0123] As can be seen, in this embodiment, the image processing device can calculate the difference between each column of packet data and the reference data, and determine whether each component in the difference is not less than a preset threshold. The preset threshold is determined based on the value of the reference data. If there is a target component not less than the preset threshold, it is determined whether the number of the target component meets a preset quantity condition. This preset quantity condition is determined based on the minimum number of pixels corresponding to a pixel in a column when the actual packet is imaged in the detector. If this condition is met, the column of packet data is determined to be packet data. Since the image processing device can determine whether each component in the difference between each column of packet data and the reference data is not less than the preset threshold, and further determine whether the number of the target component meets the preset quantity condition when there is a target component not less than the preset threshold, thus determining the packet data in the packet data, noise interference or error can be reduced, accurately determining the packet data in the packet data, thereby improving the accuracy of packet image segmentation and reducing the loss of packet data.

[0124] As one implementation method of this application, such as Figure 5As shown, the step of determining whether the number of the target component meets the preset quantity condition may include:

[0125] S501, determine whether the total number of the target components reaches a first preset number; and determine whether the total number of consecutive components in the target components reaches a second preset number; if the total number of the target components reaches the first preset number and the total number of consecutive components reaches the second preset number, execute step S502; if the total number of the target components does not reach the first preset number or the total number of consecutive components does not reach the second preset number, execute step S503.

[0126] The continuous component refers to the component in the packet data where the corresponding pixel points are consecutive pixels.

[0127] S502, confirm that the data in this column is package data;

[0128] S503, confirm that the data in this column is not parcel data.

[0129] Since the target component may be all components in the difference, may be continuous components at intervals, or may be a single component at intervals, the preset quantity condition can be determined based on the minimum number of pixels in a column of pixels when the actual component is wrapped in the detector for imaging. Then, it can be determined whether the number of target components in the difference meets the preset quantity condition. The preset quantity condition may include multiple conditions.

[0130] In one implementation, to reduce noise or errors caused by environmental factors in the overlay data, the image processing device can determine whether the total number of target components reaches a first preset number; and determine whether the total number of consecutive components in the target components reaches a second preset number, to further determine whether the column of overlay data is package data. If the total number of target components reaches the first preset number and the total number of consecutive components reaches the second preset number, then the column of overlay data meets the condition of being package data, and it can be determined that the column of overlay data is package data; otherwise, it is determined that the column of overlay data is not package data.

[0131] The first and second preset quantities can be set according to the number of pixels per column that the detector can detect. For example, if the number of pixels per column that the detector can detect is N, the first preset quantity can be 70% * N, and the second preset quantity can be 20 * N. The first and second preset quantities can also be set according to actual conditions, and no specific limitation is made here.

[0132] For example, if the detector detects 100 pixels per column, the difference between the over-packet data and the reference data is a vector. Each component in the equation has a component that is not less than a preset threshold, and these components are s1 to s2. 30 s 40 To s 60 s 70 To s 90 The first preset quantity is 50, and the second preset quantity is 30. The image processing device can determine s1 to s2. 30 s 40 To s 60 s 70 To s 90 Whether the total number reaches 50, due to s1 to s 30 s 40 To s 60 s 70 To s 90 The total number is 70, which means the first preset number has been reached, so it can be determined that s1 to s 30 s 40 To s 60 s 70 To s 90 Whether the sum of the number of consecutive components in s1 to s2 reaches the second preset number, since s1 to s2 30 s 40 To s 60 s 70 To s 90 The total number of consecutive components is 30, 20, and 20, which is the second preset number. Since both of the above conditions are met, it can be determined that the data in this column is package data.

[0133] As can be seen, in this embodiment, it is determined whether the total number of target components reaches a first preset number, and whether the total number of consecutive components in the target components reaches a second preset number. Here, a consecutive component is a component whose corresponding pixel points in the packet data are consecutive. If the total number of target components reaches the first preset number, and the total number of consecutive components reaches the second preset number, then the packet data column is determined to be packet data. Since, when the existence of target components is confirmed, further determining whether the total number of target components reaches the first preset number, and whether the total number of consecutive components in the target components reaches the second preset number, can reduce noise interference or error effects, accurately determine packet data in the packet data, thereby improving the accuracy of packet image segmentation and reducing packet data loss.

[0134] As one embodiment of this application, the above method may further include:

[0135] If there are no packages on the conveyor belt of the security inspection equipment, a series of data detected by the linear array detector is acquired as the reference data; or,

[0136] If there are no packages on the conveyor belt of the security inspection equipment, multiple columns of data detected by the linear array detector are acquired, and the average value of the multiple columns of data is calculated as the reference data.

[0137] Since the data detected by the linear detector is array data, the image processing device can determine the reference data based on the data in each column of the array data.

[0138] In one implementation, the image processing device can acquire a series of data detected by the linear array detector as reference data, even when there are no packages on the conveyor belt of the security inspection equipment.

[0139] For example, the array data detected by the linear array detector is Therefore, the image processing device can obtain any column of data as reference data, and thus obtain {n}. 01 n 02 n 03 n 04 …n 0k …n 0N} can also be used as reference data to obtain {n k1 n k2 n k3 n k4 …n kk …n kN This is for reference only.

[0140] In another implementation, due to fluctuations in the detector's radiation source, the data detected by the detector may be biased, specifically, the column data of the array may be biased. Therefore, to reduce the bias in the reference data, the image processing equipment can acquire multiple columns of data detected by the linear array detector even when there are no packages on the conveyor belt of the security inspection equipment, and calculate the average value of the multiple columns of data as reference data.

[0141] For example, the array data detected by the linear array detector is Therefore, image processing devices can acquire multiple columns of data {n 01 n 02 n 03 n 04 …n 0k …n 0N}、{n 11 n 12 n 13 n 14 …n 1k …n 1N}、{n k1 n k2 nk3 n k4 …n kk …n kN}, calculate {n 01 n 02 n 03 n 04 …n 0k …n 0N}、{n 11 n 12 n 13 n 14 …n 1k …n 1N}、{n k1 n k2 n k3 n k4 …n kk …n kN The average value of} is then used as a reference data.

[0142] As can be seen, in this embodiment, when there are no packages on the conveyor belt of the security inspection equipment, a single column of data detected by the linear array detector is acquired as reference data; or, when there are no packages on the conveyor belt of the security inspection equipment, multiple columns of data detected by the linear array detector are acquired, and the average value of the multiple columns of data is calculated as reference data. Since the reference data can be determined based on each column of data in the array data, making the reference data more accurate and reliable, the package data in the package passing data can be accurately determined based on the difference between the package passing data detected by the detector and the reference data, thereby improving the accuracy of package image segmentation and reducing the loss of package data.

[0143] As one embodiment of this application, the above method may further include:

[0144] If the detection result indicates that there is a package on the conveyor belt, the package data is determined to be package data.

[0145] Since the detection unit's detection results can include two scenarios: one where the detection result indicates that there are no packages on the conveyor belt, and another where the detection result indicates that there are packages on the conveyor belt, the image processing device can determine that the package data is package data when the detection result indicates that there are packages on the conveyor belt.

[0146] In one implementation, when the detection unit detects a package, it can mark the package with a package segmentation interruption signal. Then, the image processing device can obtain the package segmentation interruption signal in the detection result, and when it obtains the package passing data, it can determine the package passing data with the package segmentation interruption signal as package data.

[0147] For example, if the package placed on the conveyor belt of the security inspection equipment is a chair, the detection unit can mark the chair with a package segmentation interruption signal when it detects the chair. The image processing equipment can obtain the package segmentation interruption signal in the detection result. When it obtains the package data detected by the detector, it can identify the package data with the package segmentation interruption signal as chair data.

[0148] As can be seen, in this embodiment, when the detection result indicates the presence of a package on the conveyor belt, the image processing device can determine that the package data is package data. Therefore, the accuracy of package image segmentation can be improved, and the loss of package data can be reduced.

[0149] As one embodiment of this application, since the coverage of a group of detectors is limited, multiple groups of detectors can be set to increase the coverage of detection. Therefore, the above-mentioned detection unit may include multiple groups of detectors. The position of each detector in the multiple groups of detectors can be reasonably set according to the actual situation of the package being detected, so that the detection range of the multiple groups of detectors can cover the space at a preset height perpendicular to the conveyor belt transmission direction.

[0150] The step of determining the package data as package data when the detection result indicates that a package exists on the conveyor belt may include:

[0151] If the detection result of at least one of the multiple sets of detectors indicates that there is a package on the conveyor belt, the package data is determined to be package data.

[0152] Since the detection range of multiple detectors can cover a space at a preset height perpendicular to the conveyor belt's transmission direction, each detector needs to be reasonably set within the preset height range. For example, they can be set at equal intervals within the preset height range, or they can be set according to a certain ratio. No specific limitation is made here.

[0153] When a package is present on the conveyor belt, any set of detectors in the detection unit can be triggered to identify the presence of a package on the conveyor belt. Therefore, if at least one set of detectors indicates the presence of a package on the conveyor belt, the image processing device can determine that the package data is package data.

[0154] For example, such as Figure 6As shown, the detection unit includes two sets of detectors. The first set of detectors consists of a transmitter 601, a receiver 602, and a receiver 603, while the second set consists of a transmitter 604, a receiver 605, and a receiver 606. These are installed at preset heights, covering a space perpendicular to the conveyor belt's transmission direction at a preset height. When a package 607 is present on the conveyor belt, both the first and second sets of detectors are triggered, indicating the presence of the package 607. The image processing device can then determine that the package data is package data based on the detection result of either the first or second set of detectors.

[0155] As can be seen, in this embodiment, when the detection result of at least one of the multiple detectors indicates the presence of a package on the conveyor belt, the image processing device can determine that the package data is package data. Since the detection unit can include multiple detectors, the detection range can cover a space at a preset height perpendicular to the conveyor belt's transmission direction. This reduces the probability of the detection unit not being triggered when there are irregularly shaped packages on the conveyor belt. Therefore, the presence of packages on the conveyor belt can be accurately identified, thereby improving the accuracy of package image segmentation and reducing the loss of package data.

[0156] As one embodiment of this application, the aforementioned multiple detector groups include at least a first detector group and a second detector group. The first detector group includes a first transmitter, a first receiver, and a second receiver, while the second detector group includes a second transmitter, a third receiver, and a fourth receiver. The number of detector groups can be flexibly set, and the transmitters and receivers included in each detector group can be the same or different.

[0157] The first transmitter and the second transmitter are respectively disposed on both sides along the width direction of the conveyor belt, and the first receiver and the second receiver are located on the opposite side of the first transmitter. The first receiver and the second receiver are disposed at different heights perpendicular to the conveyor belt transmission direction. The first receiver and the second receiver are used to receive the detection signal emitted by the first transmitter.

[0158] The third and fourth receivers are located on the opposite side of the second transmitter. The third and fourth receivers are positioned at different heights perpendicular to the conveyor belt's transmission direction and are respectively used to receive the detection signals emitted by the second transmitter.

[0159] The detection range of the first group of detectors does not completely overlap with the detection range of the second group of detectors.

[0160] To expand the coverage area, the detection unit may include multiple sets of detectors. In one embodiment, the detection unit includes at least two sets of detectors: a first set of detectors and a second set of detectors. The first set of detectors may include a first transmitter, a first receiver, and a second receiver, while the second set of detectors may include a second transmitter, a third receiver, and a fourth receiver.

[0161] In one embodiment, the first transmitter and the second transmitter can be respectively disposed on opposite sides along the width direction of the conveyor belt. The first transmitter can transmit detection signals to the first receiver and the second receiver respectively. Therefore, the first receiver and the second receiver are located on opposite sides of the first transmitter. The first receiver and the second receiver can be disposed at different heights perpendicular to the conveyor belt transmission direction, and can be used to receive the detection signals transmitted by the first transmitter.

[0162] The second transmitter can transmit signals to the third and fourth receivers respectively. Therefore, the third and fourth receivers are located on the opposite side of the second transmitter. The third and fourth receivers can be set at different heights in the vertical conveyor belt transmission direction, and are used to receive the detection signals transmitted by the second transmitter respectively.

[0163] To increase the coverage of the detection unit, the overlap between the two sets of detectors can be reduced. Therefore, the detection ranges of the first and second sets of detectors do not need to completely overlap. For example, the lowest point of the preset height space covered by the detection range of the first set of detectors can be higher than the highest point of the preset height space covered by the detection range of the second set of detectors, or the lowest point of the preset height space covered by the detection range of the first set of detectors can be lower than the highest point of the preset height space covered by the detection range of the second set of detectors by a certain distance. No specific limitation is made here.

[0164] For example, such as Figure 6 As shown, the detection unit includes two sets of detectors, which can be infrared light barriers. The first set of infrared light barriers consists of a transmitter 601, a receiver 602, and a receiver 603, while the second set consists of a transmitter 604, a receiver 605, and a receiver 606. Transmitters 601 and 604 are located on opposite sides of the conveyor belt width direction.

[0165] In the first set of infrared light barriers, transmitter 601 can emit infrared rays to receiver 602 and receiver 603 respectively. Receiver 602 and receiver 603 are located on the opposite side of transmitter 601 and are set at different heights in the vertical conveyor belt transmission direction, so that they can receive the infrared rays emitted by transmitter 601.

[0166] In the second set of infrared light barriers, transmitter 604 can emit infrared rays to receiver 605 and receiver 606 respectively. Receiver 605 and receiver 606 are located on the opposite side of transmitter 604. Receiver 605 and receiver 606 are set at different heights in the vertical conveyor belt transmission direction and can receive the infrared rays emitted by transmitter 604.

[0167] To reduce the overlap between the two sets of infrared light barrier detections, the detection ranges of the first and second sets of infrared light barriers do not need to completely overlap. In the first set of infrared light barriers, transmitter 601, receiver 602, and receiver 603 can form a detection range within a preset height space, i.e., the height space between receiver 602 and receiver 603. Similarly, in the second set of infrared light barriers, transmitter 604, receiver 605, and receiver 606 can form a detection range within a preset height space, i.e., the height space between receiver 605 and receiver 606. Receiver 602 in the first set of infrared light barriers is higher than receiver 605 in the second set, and the detection ranges between receiver 603 and receiver 605 in the first and second sets of infrared light barriers overlap. Therefore, the detection ranges of the first and second sets of infrared light barriers do not completely overlap. When a package 607 is present on the conveyor belt, both the first and second sets of infrared light barriers will be triggered, indicating the presence of a package on the conveyor belt. The infrared light barrier triggering time is no less than 3ms, in order to perform anti-jitter processing on the infrared light barrier triggering time.

[0168] As can be seen, in this embodiment, the multiple detectors include at least a first set of detectors and a second set of detectors. The first set of detectors includes a first transmitter, a first receiver, and a second receiver. The second set of detectors includes a second transmitter, a third receiver, and a fourth receiver. The first and second transmitters are respectively disposed on opposite sides along the width of the conveyor belt. The first and second receivers are located opposite the first transmitter and are positioned at different heights perpendicular to the conveyor belt's transmission direction. The first and second receivers are used to receive the detection signal emitted by the first transmitter. The third and fourth receivers are located opposite the second transmitter and are positioned at different heights perpendicular to the conveyor belt's transmission direction, respectively, and are used to receive the detection signal emitted by the second transmitter. The detection range of the first set of detectors does not completely overlap with the detection range of the second set of detectors. By reasonably setting up the multiple sets of detectors, the detection range of the detection unit can be increased. When a package is present on the conveyor belt, the detection unit can be triggered, reducing the probability of irregularly shaped packages not being triggered, thus improving the accuracy of package image segmentation and reducing the loss of package data.

[0169] As one embodiment of this application, the step of segmenting the package data based on the package data to obtain a package image may include:

[0170] The consecutive parcel data in the parcel passing data are segmented into the same parcel image.

[0171] A package image is composed of continuous pixels. After the image processing device determines the package data in the package data, it can segment the continuous package data in the package data into the same package image, thereby obtaining the package image of the package.

[0172] For example, such as Figure 7 As shown, the package is a chair with a rectangular backrest that is hollow in the middle. The seat and legs are connected by a liftable rod. Because the backrest is hollow and the liftable rod is thin, the detection unit's detection range does not cover the backrest and liftable rod. The liftable rod portion corresponding to black line 701 and the backrest portion corresponding to black line 702 are the parts not detected by the detection unit. The image processing device can determine the package data in the package data based on the difference between the package data and the reference data. That is, it can determine the liftable rod portion corresponding to black line 701 and the backrest portion corresponding to black line 702 as package data. The continuous package data in the package data is segmented into the same package image, thus obtaining a complete chair image.

[0173] As can be seen, in this embodiment, consecutive parcel data within the parcel passing data are segmented into the same parcel image. This results in a complete parcel image, thereby improving the accuracy of parcel image segmentation and reducing the loss of parcel data.

[0174] Figure 8 This is a specific flowchart of a package image segmentation method provided in an embodiment of this application. The following is in conjunction with... Figure 8 The package image segmentation method provided in the embodiments of this application will be described with examples. For instance... Figure 8 As shown, the package image segmentation method provided in this application embodiment may include the following steps:

[0175] S801, obtains real-time air quality values ​​as reference data;

[0176] Since the air value refers to the data detected by the detector when there are no packages in the security checkpoint, the image processing equipment can obtain the air value as reference data. To prevent fluctuations in the detector's radiation source, the air value can be acquired in real time to obtain accurate and reliable reference data.

[0177] S802, retrieve packet data;

[0178] Image processing equipment can acquire packet data when the detector detects packet data.

[0179] S803, calculate the difference between the packet data and the reference data;

[0180] After the image processing device acquires the packet data, it can calculate the difference between the packet data and the reference data. Based on this difference, it can determine whether there is packet data in the packet data.

[0181] S804, Determine whether the difference meets the threshold; if it does, proceed to step S805; if it does not, proceed to step S807.

[0182] If there is package data in the packet data, the difference between the packet data and the reference data must meet the following conditions: each component in the difference is not less than a preset threshold, and there are components in the difference that are not less than the preset threshold that can be used as target components; the total number of target components reaches a first preset number, and the total number of consecutive components in the target components reaches a second preset number.

[0183] S805, marking the actual package data in the package data;

[0184] The image processing device determines whether the difference meets the threshold, which can determine whether each column of data passing through the package is package data. Then, each column of data passing through the package that is determined to be package data can be marked as actual package data.

[0185] S806, segment the package image based on the tags in the package data;

[0186] Because the package data in the package passing data is marked, the continuous package passing data in the package passing data can be segmented into the same package image based on the marking of the package data, and then the package image can be displayed.

[0187] S807, do not mark packet data.

[0188] As can be seen, in the solution provided in this application embodiment, the image processing device is applied to the security inspection system. The security inspection system also includes security inspection equipment, which has detectors and detection units. The detection unit includes multiple sets of detectors. The detection range of the multiple sets of detectors covers a space at a preset height perpendicular to the conveyor belt transmission direction of the security inspection equipment. The image processing device can acquire the baggage data detected by the detectors and the detection results of the detection units during the operation of the security inspection equipment. The detection unit is used to detect whether there are packages on the conveyor belt of the security inspection equipment. When the detection result indicates that there are no packages on the conveyor belt, the package data in the baggage data is determined based on the difference between the baggage data and the reference data. The reference data is the data detected by the detectors when there are no packages on the conveyor belt, which is obtained in advance. The package data refers to the detection data in the baggage data that is judged to belong to packages. Based on the package data, the baggage data is segmented to obtain a package image. Since the detection unit detects no packages on the conveyor belt, the image processing device can further determine the package data in the package passing data based on the difference between the package passing data detected by the detector and the reference data. This allows the package passing data to be segmented to obtain a package image. In this way, even for irregularly shaped packages or other packages that may cause inaccurate detection results from the detection unit, the package data in the package passing data can be accurately determined. This enables the package data corresponding to a single package to be accurately segmented into a single package image, improving the accuracy of package image segmentation and reducing the loss of package data.

[0189] Figure 9 This is another specific flowchart of the package image segmentation method provided in the embodiments of this application. The following is in conjunction with... Figure 9 The package image segmentation method provided in the embodiments of this application will be described with examples. For instance... Figure 9 As shown, the package image segmentation method provided in this application embodiment may include the following steps:

[0190] S901, There are actual packages on the conveyor belt;

[0191] When there is an actual package on the conveyor belt, the detection unit of the security inspection equipment can detect it. However, for packages with extremely irregular shapes, the detection unit may fail to detect the package.

[0192] S902, the detector obtains data;

[0193] When a package travels along the conveyor belt into the detection range of the security screening equipment, the detector can capture the data, known as package transit data. Image processing equipment can then acquire this package transit data.

[0194] S903, Does the difference between the packet data and the air value meet the threshold?

[0195] After acquiring the packet data, the image processing device can calculate the difference between the packet data and the reference data, determine whether the difference meets the threshold, and then determine whether there is packet data in the packet data and mark the packet data.

[0196] S904, acquire the light barrier trigger signal;

[0197] When the detection unit detects an actual package on the conveyor belt, the package can be marked. For example, the detection unit includes multiple sets of infrared light barriers. When a package is present on the conveyor belt, at least one set of infrared light barriers will be triggered, the image processing device can acquire the light barrier trigger signal, and the package will have a package segmentation interruption signal.

[0198] S905, segment the image based on the tags in the package data;

[0199] Image processing equipment can identify package data based on its markings and segment the package data into a single package image. Identifying package data involves two steps: first, determining if there is a package segmentation interruption signal. The presence of an interruption signal indicates that the passed data is package data, which can be directly marked as such. Second, if no interruption signal is present, the package data is identified based on the difference between the passed data and reference data, provided the difference meets a threshold. Consecutive passed data that are package data are then segmented into the same package image.

[0200] S906, Package Image Output.

[0201] Display the segmented package image.

[0202] As can be seen, in the solution provided in this application embodiment, the image processing device is applied to the security inspection system. The security inspection system also includes security inspection equipment, which has detectors and detection units. The detection unit includes multiple sets of detectors. The detection range of the multiple sets of detectors covers a space at a preset height perpendicular to the conveyor belt transmission direction of the security inspection equipment. The image processing device can acquire the baggage data detected by the detectors and the detection results of the detection units during the operation of the security inspection equipment. The detection unit is used to detect whether there are packages on the conveyor belt of the security inspection equipment. When the detection result indicates that there are no packages on the conveyor belt, the package data in the baggage data is determined based on the difference between the baggage data and the reference data. The reference data is the data detected by the detectors when there are no packages on the conveyor belt, which is obtained in advance. The package data refers to the detection data in the baggage data that is judged to belong to packages. Based on the package data, the baggage data is segmented to obtain a package image. Since the detection unit detects no packages on the conveyor belt, the image processing device can further determine the package data in the package passing data based on the difference between the package passing data detected by the detector and the reference data. This allows the package passing data to be segmented to obtain a package image. In this way, even for irregularly shaped packages or other packages that may cause inaccurate detection results from the detection unit, the package data in the package passing data can be accurately determined. This enables the package data corresponding to a single package to be accurately segmented into a single package image, improving the accuracy of package image segmentation and reducing the loss of package data.

[0203] Corresponding to the above-mentioned package image segmentation method, this application embodiment also provides a package image segmentation device, which will be described below.

[0204] like Figure 10 As shown, a package image segmentation device is configured in the image processing equipment of a security inspection system. The security inspection system also includes security inspection equipment, which has detectors and detection units. The detection unit includes multiple sets of detectors, and the detection range of the multiple sets of detectors covers a space at a preset height perpendicular to the conveyor belt transport direction of the security inspection equipment. The device includes:

[0205] The package data and detection result acquisition module 1010 is used to acquire the package data detected by the detector and the detection result of the detection unit during the operation of the security inspection equipment, wherein the detection unit is used to detect whether there is a package on the conveyor belt of the security inspection equipment;

[0206] The first determining module 1020 is used to determine package data in the package passing data based on the difference between the package passing data and the reference data when the detection result indicates that there are no packages on the conveyor belt. The reference data is the data detected by the detector when there are no packages on the conveyor belt, which is obtained in advance. The package data refers to the detection data in the package passing data that is judged to belong to a package.

[0207] The package image segmentation module 1030 is used to segment the package data based on the package data to obtain a package image.

[0208] As can be seen, in the solution provided in this application embodiment, the image processing device is applied to the security inspection system. The security inspection system also includes security inspection equipment, which has detectors and detection units. The detection unit includes multiple sets of detectors. The detection range of the multiple sets of detectors covers a space at a preset height perpendicular to the conveyor belt transmission direction of the security inspection equipment. The image processing device can acquire the baggage data detected by the detectors and the detection results of the detection units during the operation of the security inspection equipment. The detection unit is used to detect whether there are packages on the conveyor belt of the security inspection equipment. When the detection result indicates that there are no packages on the conveyor belt, the package data in the baggage data is determined based on the difference between the baggage data and the reference data. The reference data is the data detected by the detectors when there are no packages on the conveyor belt, which is obtained in advance. The package data refers to the detection data in the baggage data that is judged to belong to packages. Based on the package data, the baggage data is segmented to obtain a package image. Since the detection unit detects no packages on the conveyor belt, the image processing device can further determine the package data in the package passing data based on the difference between the package passing data detected by the detector and the reference data. This allows the package passing data to be segmented to obtain a package image. In this way, even for irregularly shaped packages or other packages that may cause inaccurate detection results from the detection unit, the package data in the package passing data can be accurately determined. This enables the package data corresponding to a single package to be accurately segmented into a single package image, improving the accuracy of package image segmentation and reducing the loss of package data.

[0209] In one embodiment of this application, the detector is a linear array detector; the first determining module 1020 may include:

[0210] The preset threshold judgment submodule is used to calculate the difference between each column of over-packed data and the reference data, and to determine whether each component of the difference is not less than a preset threshold, wherein the preset threshold is determined based on the value of the reference data;

[0211] The preset quantity condition judgment submodule is used to determine whether the quantity of the target component satisfies the preset quantity condition if there is a target component that is not less than the preset threshold. The preset quantity condition is determined based on the minimum number of pixels in a column of pixels when the actual image is encapsulated in the detector.

[0212] The first determination submodule is used to determine if the data in the column is package data if certain conditions are met.

[0213] As one embodiment of this application, the preset quantity condition judgment submodule may include:

[0214] A preset quantity judgment unit is used to determine whether the total number of target components in the difference reaches a first preset quantity; and to determine whether the total number of continuous components in the target components reaches a second preset quantity, wherein the continuous component is a component whose corresponding pixel point in the packet data is a continuous pixel point;

[0215] The package data determination unit is used to determine that the column of package data is package data if the total number of the target components reaches a first preset number and the total number of the consecutive components reaches a second preset number.

[0216] As one embodiment of this application, the apparatus may further include:

[0217] The reference data acquisition module is used to acquire a series of data detected by the linear array detector as the reference data when there are no packages on the conveyor belt of the security inspection equipment; or,

[0218] This is used to acquire multiple columns of data detected by the linear array detector when there are no packages on the conveyor belt of the security inspection equipment, and to calculate the average value of the multiple columns of data as the reference data.

[0219] As one embodiment of this application, the apparatus may further include:

[0220] The second determining module is used to determine that the package passing data is package data when the detection results of at least one of the multiple sets of detectors indicate that there is a package on the conveyor belt.

[0221] As one embodiment of this application, the multiple sets of detectors include at least a first set of detectors and a second set of detectors. The first set of detectors includes a first transmitter, a first receiver, and a second receiver. The second set of detectors includes a second transmitter, a third receiver, and a fourth receiver.

[0222] The first transmitter and the second transmitter are respectively disposed on both sides along the width direction of the conveyor belt, and the first receiver and the second receiver are located on the opposite side of the first transmitter. The first receiver and the second receiver are disposed at different heights perpendicular to the conveyor belt transmission direction. The first receiver and the second receiver are used to receive the detection signal emitted by the first transmitter.

[0223] The third receiver and the fourth receiver are located on the opposite side of the second transmitter. The third receiver and the fourth receiver are set at different heights perpendicular to the conveyor belt transmission direction, and are respectively used to receive the detection signal emitted by the second transmitter.

[0224] The detection range of the first group of detectors does not completely overlap with the detection range of the second group of detectors.

[0225] As one embodiment of this application, the package image segmentation module 1030 may include:

[0226] The package image segmentation submodule is used to segment consecutive package data in the package data into the same package image.

[0227] This application also provides a security inspection system, such as... Figure 11 As shown, the security inspection system includes an image processing device 1110 and a security inspection device 1120. The security inspection device has a detector 1121 and a detection unit 1122. The detection unit includes multiple sets of detectors 1123. The detection range of the multiple sets of detectors 1123 covers a space at a preset height perpendicular to the conveyor belt transport direction of the security inspection device, wherein:

[0228] The detection unit 1122 is used to detect whether there are packages on the conveyor belt of the security inspection equipment and obtain the detection result;

[0229] The detector 1121 is used to detect packages on the conveyor belt of the security inspection equipment during the operation of the security inspection equipment and obtain package data.

[0230] The image processing device 1110 is used to acquire the package passing data and the detection result; when the detection result indicates that there is no package on the conveyor belt, it determines the package data in the package passing data based on the difference between the package passing data and the reference data; based on the package data, it segments the package passing data to obtain a package image, wherein the reference data is the data detected by the detector when there is no package on the conveyor belt, and the package data refers to the detection data in the package passing data that is judged to belong to a package.

[0231] As can be seen, in the solution provided in this application embodiment, the image processing device is applied to the security inspection system. The security inspection system also includes security inspection equipment, which has detectors and detection units. The detection unit includes multiple sets of detectors. The detection range of the multiple sets of detectors covers a space at a preset height perpendicular to the conveyor belt transmission direction of the security inspection equipment. The image processing device can acquire the baggage data detected by the detectors and the detection results of the detection units during the operation of the security inspection equipment. The detection unit is used to detect whether there are packages on the conveyor belt of the security inspection equipment. When the detection result indicates that there are no packages on the conveyor belt, the package data in the baggage data is determined based on the difference between the baggage data and the reference data. The reference data is the data detected by the detectors when there are no packages on the conveyor belt, which is obtained in advance. The package data refers to the detection data in the baggage data that is judged to belong to packages. Based on the package data, the baggage data is segmented to obtain a package image. Since the detection unit detects no packages on the conveyor belt, the image processing device can further determine the package data in the package passing data based on the difference between the package passing data detected by the detector and the reference data. This allows the package passing data to be segmented to obtain a package image. In this way, even for irregularly shaped packages or other packages that may cause inaccurate detection results from the detection unit, the package data in the package passing data can be accurately determined. This enables the package data corresponding to a single package to be accurately segmented into a single package image, improving the accuracy of package image segmentation and reducing the loss of package data.

[0232] As one embodiment of this application, the detector is a linear array detector;

[0233] The image processing device 1110 can specifically be used to calculate the difference between each column of overpacking data and reference data, determine whether each component in the difference is not less than a preset threshold, wherein the preset threshold is determined based on the value of the reference data; if there is a target component not less than the preset threshold, determine whether the number of the target component meets a preset quantity condition, wherein the preset quantity condition is determined based on the minimum number of pixels corresponding to a column of pixels when the actual package is imaged in the detector; if it meets the condition, determine that the column of overpacking data is package data.

[0234] As one embodiment of this application, the image processing device 1110 can be specifically used to determine whether the total number of target components in the difference reaches a first preset number; and to determine whether the total number of continuous components in the target components reaches a second preset number, wherein the continuous component is the component whose corresponding pixel point in the packet data is a continuous pixel point; if the total number of target components reaches the first preset number and the total number of continuous components reaches the second preset number, the packet data column is determined to be packet data.

[0235] As one embodiment of this application, the image processing device 1110 can also be used to acquire a column of data detected by the linear array detector as the reference data when there are no packages on the conveyor belt of the security inspection equipment; or, it can also be used to acquire multiple columns of data detected by the linear array detector and calculate the average value of the multiple columns of data as the reference data when there are no packages on the conveyor belt of the security inspection equipment.

[0236] As one embodiment of this application, the image processing device 1110 is capable of determining that the package passing data is package data when the detection result of at least one of the multiple sets of detectors indicates that there is a package on the conveyor belt.

[0237] As one embodiment of this application, the multiple sets of detectors 1123 include at least a first set of detectors and a second set of detectors. The first set of detectors includes a first transmitter, a first receiver, and a second receiver. The second set of detectors includes a second transmitter, a third receiver, and a fourth receiver.

[0238] The first transmitter and the second transmitter are respectively disposed on both sides along the width direction of the conveyor belt, and the first receiver and the second receiver are located on the opposite side of the first transmitter. The first receiver and the second receiver are disposed at different heights perpendicular to the conveyor belt transmission direction. The first receiver and the second receiver are used to receive the detection signal emitted by the first transmitter.

[0239] The third receiver and the fourth receiver are located on the opposite side of the second transmitter. The third receiver and the fourth receiver are set at different heights perpendicular to the conveyor belt transmission direction, and are respectively used to receive the detection signal emitted by the second transmitter.

[0240] The detection range of the first group of detectors does not completely overlap with the detection range of the second group of detectors.

[0241] As one embodiment of this application, the image processing device 1110 can be specifically used to segment the continuous package data in the package data into the same package image.

[0242] This application embodiment also provides an image processing device, which is an image processing device in a security inspection system. The security inspection system further includes security inspection equipment, which has a detector and a detector, such as... Figure 12 As shown, the image processing device includes:

[0243] Memory 1201 is used to store computer programs;

[0244] When the processor 1202 executes the program stored in the memory 1201, it implements the steps of the package image segmentation method described in any of the above embodiments.

[0245] Furthermore, the aforementioned image processing device may also include a communication bus and / or a communication interface, with the processor 1202, the communication interface, and the memory 1201 communicating with each other via the communication bus.

[0246] As can be seen, in the solution provided in this application embodiment, the image processing device is applied to the security inspection system. The security inspection system also includes security inspection equipment, which has detectors and detection units. The detection unit includes multiple sets of detectors. The detection range of the multiple sets of detectors covers a space at a preset height perpendicular to the conveyor belt transmission direction of the security inspection equipment. The image processing device can acquire the baggage data detected by the detectors and the detection results of the detection units during the operation of the security inspection equipment. The detection unit is used to detect whether there are packages on the conveyor belt of the security inspection equipment. When the detection result indicates that there are no packages on the conveyor belt, the package data in the baggage data is determined based on the difference between the baggage data and the reference data. The reference data is the data detected by the detectors when there are no packages on the conveyor belt, which is obtained in advance. The package data refers to the detection data in the baggage data that is judged to belong to packages. Based on the package data, the baggage data is segmented to obtain a package image. Since the detection unit detects no packages on the conveyor belt, the image processing device can further determine the package data in the package passing data based on the difference between the package passing data detected by the detector and the reference data. This allows the package passing data to be segmented to obtain a package image. In this way, even for irregularly shaped packages or other packages that may cause inaccurate detection results from the detection unit, the package data in the package passing data can be accurately determined. This enables the package data corresponding to a single package to be accurately segmented into a single package image, improving the accuracy of package image segmentation and reducing the loss of package data.

[0247] The communication bus mentioned in the image processing device above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not indicate that there is only one bus or one type of bus.

[0248] The communication interface is used for communication between the aforementioned image processing device and other devices.

[0249] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0250] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0251] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the steps of the package image segmentation method described in any of the above embodiments.

[0252] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to perform the package image segmentation method described in any of the embodiments above.

[0253] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0254] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0255] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for apparatuses, systems, image processing devices, computer-readable storage media, and computer program products, since they are basically similar to the method embodiments, the descriptions are relatively simple, and relevant parts can be referred to the descriptions of the method embodiments.

[0256] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. A parcel image segmentation method, characterized by, The application discloses an image processing device applied to a security inspection system, wherein the security inspection system further comprises a security inspection device, the security inspection device has a detector and a detection unit, the detection unit comprises a plurality of groups of detectors, the detection ranges of the plurality of groups of detectors cover the space of a preset height perpendicular to the conveying belt transmission direction of the security inspection device, and the method comprises the following steps: obtaining package data detected by the detector during the operation of the security inspection device and a detection result of the detection unit, wherein the detection unit is used for detecting whether there is a package on the conveying belt of the security inspection device; in the case that the detection result indicates that there is no package on the conveying belt, determining package data in the package data based on the difference between the package data and reference data, wherein the reference data is data detected by the detector in the case that there is no package on the conveying belt and is obtained in advance, and the package data refers to detection data in the package data which is judged to belong to a package; based on the package data, the package data is segmented to obtain a package image.

2. The method of claim 1, wherein, The detector is a linear array detector; the step of determining the package data in the package data based on the difference between the package data and the reference data comprises the following steps: calculating the difference value between each column of package data and the reference data, and judging whether each component in the difference value is not less than a preset threshold value, wherein the preset threshold value is determined based on the numerical value of the reference data; if there is a target component which is not less than the preset threshold value, judging whether the number of the target component meets a preset number condition, wherein the preset number condition is determined based on the minimum pixel number corresponding to an actual package when the actual package is imaged in one column of pixels in the detector; if yes, determining that the column of package data is package data.

3. The method of claim 2, wherein, The step of judging whether the number of the target component meets the preset number condition comprises the following steps: judging whether the total number of target components in the difference value reaches a first preset number, and judging whether the total number of continuous components in the target components reaches a second preset number, wherein the continuous components are components in which the corresponding pixel points in the package data are continuous pixel points; if the total number of the target components reaches the first preset number and the total number of the continuous components reaches the second preset number, it is determined that the column of package data is package data.

4. The method of claim 2, wherein, The method further comprises the following steps: in the case that there is no package on the conveying belt of the security inspection device, obtaining one column of data detected by the linear array detector as the reference data; or in the case that there is no package on the conveying belt of the security inspection device, obtaining a plurality of columns of data detected by the linear array detector, calculating the average value of the plurality of columns of data as the reference data.

5. The method according to any one of claims 1 to 4, characterized in that, The method further comprises the following steps: in the case that the detection result of at least one group of detectors in the plurality of groups of detectors indicates that there is a package on the conveying belt, it is determined that the package data is package data.

6. The method of claim 1, wherein, the plurality of groups of detectors at least comprise a first group of detectors and a second group of detectors, the first group of detectors comprise a first transmitter, a first receiver and a second receiver, and the second group of detectors comprise a second transmitter, a third receiver and a fourth receiver; The first transmitter and the second transmitter are respectively arranged on two sides along the width direction of the conveying belt, the first receiver and the second receiver are located on the opposite side of the first transmitter, and the first receiver and the second receiver are arranged at positions of different heights in the direction perpendicular to the conveying direction of the conveying belt and are used for receiving the detection signals transmitted by the first transmitter. The third receiver and the fourth receiver are located on the opposite side of the second transmitter, and the third receiver and the fourth receiver are arranged at positions of different heights in the direction perpendicular to the conveying direction of the conveying belt and are respectively used for receiving the detection signals transmitted by the second transmitter. The detection range of the first group of detectors and the detection range of the second group of detectors do not completely overlap.

7. The method according to any one of claims 1 to 4, characterized in that, The step of segmenting the over-package data based on the package data to obtain the package image comprises: The over-package data that is continuously package data is segmented into the same package image.

8. A parcel image segmentation apparatus, characterized by, The device is configured in an image processing equipment in a security inspection system, and the security inspection system further comprises a security inspection equipment, the security inspection equipment has a detector and a detection unit, the detection unit comprises a plurality of groups of detectors, the detection ranges of the plurality of groups of detectors cover a space of a preset height in a direction perpendicular to the conveying direction of a conveying belt of the security inspection equipment, and the device comprises: An over-package data and detection result acquisition module is configured to acquire over-package data detected by the detector during operation of the security inspection equipment and detection results of the detection unit, wherein the detection unit is configured to detect whether there is a package on the conveying belt of the security inspection equipment. A first determination module is configured to, in a case where the detection result indicates that there is no package on the conveying belt, determine package data in the over-package data based on a difference between the over-package data and reference data, wherein the reference data is data detected by the detector in a case where there is no package on the conveying belt and is acquired in advance, and the package data refers to detection data in the over-package data that is judged to belong to a package. A package image segmentation module is configured to segment the over-package data based on the package data to obtain a package image.

9. A security system, characterized by The security inspection system comprises an image processing equipment and a security inspection equipment, the security inspection equipment has a detector and a detection unit, the detection unit comprises a plurality of groups of detectors, and the detection ranges of the plurality of groups of detectors cover a space of a preset height in a direction perpendicular to the conveying direction of a conveying belt of the security inspection equipment, wherein: The detection unit is configured to detect whether there is a package on the conveying belt of the security inspection equipment to obtain a detection result. The detector is configured to detect a package on the conveying belt of the security inspection equipment during operation of the security inspection equipment to obtain over-package data. The image processing device is configured to: acquire the over package data and the detection result; in a case where the detection result indicates that there is no package on the conveying belt, determine package data in the over package data based on a difference between the over package data and reference data; and segment the over package data based on the package data to obtain a package image, wherein the reference data is data detected by the detector in a case where there is no package on the conveying belt and the package data refers to detection data in the over package data that is determined to belong to a package.

10. An image processing apparatus characterized by comprising: The application is applied to a security inspection system, and the security inspection system further comprises a security inspection device having a detector and a detection unit, the detection unit comprises a plurality of groups of detectors, detection ranges of the plurality of groups of detectors cover a space of a preset height perpendicular to a conveying direction of a conveying belt of the security inspection device, and the image processing device comprises: a memory configured to store a computer program; a processor configured to execute the program stored in the memory to implement the method in any one of claims 1-7.

Images