Article marking method and device

Abstract

The embodiment of the invention provides an article marking method and device, and relates to the technical field of three-dimensional image detection. The method comprises the following steps: acquiring a three-dimensional image of a detected object, and displaying the three-dimensional image; detecting a first boundary of each object included in the detected object in the three-dimensional image of the first visual angle, and displaying the detected first boundary; in response to a confirmation operation of a user for a first target boundary of the suspicious article, in the three-dimensional image, determining a three-dimensional sub-image corresponding to the suspicious article indicated by the first target boundary at the second visual angle; detecting a second boundary of each article in the three-dimensional sub-graph of the second visual angle, and determining a second target boundary of the suspicious article based on the detected second boundary; and based on the second target boundary and the three-dimensional sub-graph, determining a three-dimensional region of the suspicious article in the three-dimensional image, and marking the three-dimensional region as the three-dimensional region of the suspicious article. By performing double boundary detection on the three-dimensional image, the efficiency and accuracy of article marking are improved.

Description

Technical Field

[0001] The present application relates to the technical field of three-dimensional image detection, and in particular to an object marking method and device. Background Art

[0002] During the security inspection process, the security inspection system detects the inspected object and displays a three-dimensional image of the inspected object. The security inspector marks the position of the suspicious item included in the inspected object in the three-dimensional image and defines the category of item to which the suspicious item belongs, and obtains the marking result of the suspicious item, so that the suspicious item can be re-inspected according to the marking result in the subsequent unpacking and re-inspection link.

[0003] In the related art, security inspectors need to manually draw the locations of suspicious objects in three-dimensional images multiple times. This operation is cumbersome and time-consuming, resulting in low object marking efficiency and, in turn, low overall security inspection efficiency. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide an object marking method and device to improve the efficiency of suspicious object detection. The specific technical solution is as follows:

[0005] In a first aspect, an embodiment of the present application provides an object marking method, the method comprising:

[0006] Acquire a three-dimensional image of the object under inspection and display the three-dimensional image;

[0007] Detecting a first boundary of each object included in the inspected object in the three-dimensional image of the first viewing angle, and displaying the detected first boundary;

[0008] In response to a user's confirmation operation on a first target boundary of a suspicious object, determining, in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at a second viewing angle; wherein the second viewing angle is different from the first viewing angle;

[0009] Detecting a second boundary of each object in the three-dimensional sub-image of the second viewing angle, and determining a second target boundary of the suspicious object based on the detected second boundary;

[0010] Based on the second object boundary and the three-dimensional sub-image, a three-dimensional region of the suspicious object in the three-dimensional image is determined and marked as the three-dimensional region of the suspicious object.

[0011] Optionally, the step of detecting a first boundary of each object included in the inspected object in the three-dimensional image of the first viewing angle and displaying the detected first boundary includes:

[0012] Acquire a first target two-dimensional image of the first viewing angle corresponding to the three-dimensional image, and perform boundary detection on the first target two-dimensional image to obtain first boundaries of each item included in the inspected object; display the detected first boundaries in the first target two-dimensional image; or, based on a conversion relationship between the first target two-dimensional image and the three-dimensional image, draw the detected first boundaries into the three-dimensional image, and display the first boundaries in the three-dimensional image;

[0013] or,

[0014] Boundary detection is performed on the three-dimensional image to obtain first boundaries of each item included in the inspected object, and the detected first boundaries are displayed in the three-dimensional image.

[0015] Optionally, in response to the user's confirmation operation on the first target boundary of the suspicious object, the step of determining, in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at the second viewing angle includes:

[0016] In the case where the first boundary is displayed in the first target two-dimensional image, in response to a selection operation of the user on the first target boundary, or in response to a drawing operation of the user on the first target boundary, based on a conversion relationship between the first target two-dimensional image and the three-dimensional image, the first target boundary is drawn into the three-dimensional image; in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at the second viewing angle is intercepted;

[0017] When the first boundary is displayed in the three-dimensional image, in response to the user's selection operation on the first target boundary, or in response to the user's drawing operation on the first target boundary, in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at the second perspective is captured.

[0018] Optionally, the step of detecting a second boundary of each object in the three-dimensional sub-image of the second viewing angle includes:

[0019] Acquire a second target two-dimensional image of the second viewing angle corresponding to the three-dimensional sub-image, and perform boundary detection on the second target two-dimensional image to obtain a second boundary of each object in the second target two-dimensional image; or,

[0020] Boundary detection is performed on the three-dimensional sub-image to obtain a second boundary of each object in the three-dimensional sub-image.

[0021] Optionally, the step of determining a second target boundary of the suspicious object based on the detected second boundary includes:

[0022] When the number of the detected second boundaries is 1, determining the detected second boundaries as the second target boundaries of the suspicious object;

[0023] In the case that the number of detected second boundaries is multiple, the second boundary indicated by the user's selection operation on the second boundary, or the boundary drawn by the user, is determined as the second target boundary of the suspicious object.

[0024] Optionally, when there are multiple detected second boundaries, before the step of determining the second boundary indicated by the user's selection operation on the second boundary, or the boundary drawn by the user, as the second target boundary of the suspicious object, the method further includes:

[0025] In the case of performing boundary detection on the second target two-dimensional image, displaying the detected second boundary in the second target two-dimensional image, or, based on a conversion relationship between the second target two-dimensional image and the three-dimensional sub-image, drawing the detected second boundary into the three-dimensional sub-image, and displaying the second boundary in the three-dimensional sub-image;

[0026] When the boundary detection is performed on the three-dimensional sub-image, the detected second boundary is displayed in the three-dimensional sub-image.

[0027] Optionally, the step of determining a three-dimensional area of ​​the suspicious object in the three-dimensional image based on the second target boundary and the three-dimensional sub-image includes:

[0028] According to the conversion relationship between the first viewing angle and the second viewing angle, a three-dimensional area corresponding to the area corresponding to the second object boundary in the three-dimensional sub-image in the three-dimensional image is determined.

[0029] Optionally, the method further includes:

[0030] Obtaining the item category of the suspicious item;

[0031] The article category of the suspicious article and the three-dimensional area of ​​the suspicious article are recorded to obtain the marking result of the inspected article so that the staff can re-inspect the suspicious article.

[0032] In a second aspect, an embodiment of the present application provides an object marking device, the device comprising:

[0033] An image display module, used to obtain a three-dimensional image of the object under inspection and display the three-dimensional image;

[0034] A first boundary detection module, used to detect a first boundary of each object included in the detected object in the three-dimensional image of the first viewing angle, and display the detected first boundary;

[0035] a three-dimensional sub-image determination module, configured to determine, in response to a user's confirmation operation on a first target boundary of a suspicious object, a three-dimensional sub-image corresponding to a second viewing angle of the suspicious object indicated by the first target boundary in the three-dimensional image; wherein the second viewing angle is different from the first viewing angle;

[0036] A second boundary detection module, configured to detect a second boundary of each object in the three-dimensional sub-image of the second viewing angle, and determine a second target boundary of the suspicious object based on the detected second boundary;

[0037] A three-dimensional region determination module is used to determine the three-dimensional region of the suspicious object in the three-dimensional image based on the second target boundary and the three-dimensional sub-image, and mark it as the three-dimensional region of the suspicious object.

[0038] Optionally, the first boundary detection module includes:

[0039] A first detection unit is used to obtain a first target two-dimensional image of the first viewing angle corresponding to the three-dimensional image, and perform boundary detection on the first target two-dimensional image to obtain a first boundary of each item included in the detected object; display the detected first boundary in the first target two-dimensional image; or, based on a conversion relationship between the first target two-dimensional image and the three-dimensional image, draw the detected first boundary into the three-dimensional image, and display the first boundary in the three-dimensional image;

[0040] or,

[0041] The second detection unit is used to perform boundary detection on the three-dimensional image to obtain a first boundary of each object included in the detected object, and display the detected first boundary in the three-dimensional image.

[0042] Optionally, the three-dimensional sub-image determination module includes:

[0043] A first sub-image determining unit is configured to, when the first boundary is displayed in the first target two-dimensional image, in response to a selection operation of the user on the first target boundary, or in response to a drawing operation of the user on the first target boundary, draw the first target boundary into the three-dimensional image based on a conversion relationship between the first target two-dimensional image and the three-dimensional image; and intercept, in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at a second viewing angle;

[0044] The second sub-image determining unit is used to intercept, in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at a second perspective in response to a selection operation of the user on the first target boundary or a drawing operation of the user on the first target boundary when the first boundary is displayed in the three-dimensional image.

[0045] Optionally, the second boundary detection module includes:

[0046] A second boundary detection unit is used to obtain a second target two-dimensional image of the second perspective corresponding to the three-dimensional sub-image, and perform boundary detection on the second target two-dimensional image to obtain a second boundary of each object in the second target two-dimensional image; or, perform boundary detection on the three-dimensional sub-image to obtain a second boundary of each object in the three-dimensional sub-image.

[0047] In a third aspect, an embodiment of the present application provides an electronic device, including:

[0048] Memory, used to store computer programs;

[0049] The processor is used to implement any method described in the first aspect when executing a program stored in the memory.

[0050] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any method described in the first aspect is implemented.

[0051] In a fifth aspect, an embodiment of the present application further provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute any of the methods described in the first aspect.

[0052] Beneficial effects of the embodiments of the present application:

[0053] In the technical solution provided by the embodiment of the present application, the electronic device can obtain a three-dimensional image of the inspected object and display the three-dimensional image; detect the first boundary of each item included in the inspected object in the three-dimensional image of the first perspective, and display the detected first boundary; in response to the user's confirmation operation on the first target boundary of the suspicious item, determine in the three-dimensional image a three-dimensional sub-graph corresponding to the suspicious item indicated by the first target boundary at the second perspective; wherein the second perspective is different from the first perspective; detect the second boundary of each item in the three-dimensional sub-graph of the second perspective, and determine the second target boundary of the suspicious item based on the detected second boundary; based on the second target boundary and the three-dimensional sub-graph, determine the three-dimensional area of ​​the suspicious item in the three-dimensional image, and mark it as the three-dimensional area of ​​the suspicious item.

[0054] It can be seen that the electronic device can determine the boundaries of suspicious items at two different perspectives by performing dual intelligent boundary detection on the three-dimensional image of the first perspective and the three-dimensional sub-image of the second perspective, avoiding the inaccurate boundary detection results caused by mutual occlusion between objects at a single perspective, improving the accuracy of the suspicious item boundary detection results, and then using the detected boundary of the suspicious item to mark the three-dimensional area of ​​the suspicious item in the three-dimensional image, improving the accuracy of the suspicious item marking results. In addition, the electronic device can perform intelligent detection on the boundary of the suspicious item by itself, and the user does not need to draw the boundary of the suspicious item multiple times, but only needs to confirm the boundary of the suspicious item detected by the electronic device to realize the marking of the suspicious item. The operation is simple and the marking process takes a short time, which improves the user's image reading efficiency and item marking efficiency, thereby improving the overall security inspection efficiency.

[0055] Of course, implementing any product or method of the present application does not necessarily require achieving all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other embodiments can also be obtained based on these drawings.

[0057] Figure 1 A flowchart of an object marking method provided in an embodiment of the present application;

[0058] Figure 2 A schematic diagram of a three-dimensional image of an object under test provided in an embodiment of the present application;

[0059] Figure 3 A schematic diagram of a three-dimensional image from a first perspective provided in an embodiment of the present application;

[0060] Figure 4 A schematic diagram of the first boundary of each object included in the object to be inspected provided in an embodiment of the present application;

[0061] Figure 5 A schematic diagram of a three-dimensional area of ​​a suspicious object in a three-dimensional image provided by an embodiment of the present application;

[0062] Figure 6 A schematic diagram of a three-dimensional sub-image of a second viewing angle provided in an embodiment of the present application;

[0063] Figure 7 Another schematic diagram of a three-dimensional sub-image of a second viewing angle provided in an embodiment of the present application;

[0064] Figure 8 A schematic diagram of a second boundary of a suspicious object provided in an embodiment of the present application;

[0065] Fig. 9 Another schematic diagram of the second boundary of the suspicious object provided in the embodiment of the present application;

[0066] Fig.10 A schematic diagram of the structure of an object marking device provided in an embodiment of the present application;

[0067] Fig.11 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0068] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field based on the present application belong to the scope of protection of the present application.

[0069] In order to improve the efficiency of item marking and thus improve the overall security inspection efficiency, the embodiments of the present application provide an item marking method, device, electronic device, computer-readable storage medium and computer program product. The following first introduces an item marking method provided in the embodiments of the present application.

[0070] The object marking method provided in the embodiment of the present application can be applied to any electronic device that can mark an object. For example, it can be applied to a processor in an X-ray detection device with a detector, including but not limited to a security inspection machine, industrial detection equipment, etc., which is not specifically limited here. For the sake of clarity, it is referred to as an electronic device in the following.

[0071] like Figure 1 As shown, a method for marking an item, the method comprising:

[0072] S101: Acquire a three-dimensional image of an object under inspection, and display the three-dimensional image.

[0073] S102: Detecting first boundaries of each object included in the inspected object in the three-dimensional image of the first viewing angle, and displaying the detected first boundaries.

[0074] S103: In response to a user's confirmation operation on a first target boundary of a suspicious object, determining, in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at a second viewing angle.

[0075] The second viewing angle is different from the first viewing angle.

[0076] S104: Detecting a second boundary of each object in the three-dimensional sub-image of the second viewing angle, and determining a second target boundary of the suspicious object based on the detected second boundary.

[0077] S105: Determine a three-dimensional region of the suspicious object in the three-dimensional image based on the second object boundary and the three-dimensional sub-image, and mark it as the three-dimensional region of the suspicious object.

[0078] In the technical solution provided by the embodiment of the present application, the electronic device can obtain a three-dimensional image of the inspected object and display the three-dimensional image; detect the first boundary of each item included in the inspected object in the three-dimensional image of the first perspective, and display the detected first boundary; in response to the user's confirmation operation on the first target boundary of the suspicious item, determine in the three-dimensional image a three-dimensional sub-graph corresponding to the suspicious item indicated by the first target boundary at the second perspective; wherein the second perspective is different from the first perspective; detect the second boundary of each item in the three-dimensional sub-graph of the second perspective, and determine the second target boundary of the suspicious item based on the detected second boundary; based on the second target boundary and the three-dimensional sub-graph, determine the three-dimensional area of ​​the suspicious item in the three-dimensional image, and mark it as the three-dimensional area of ​​the suspicious item.

[0079] It can be seen that the electronic device can determine the boundaries of suspicious items at two different perspectives by performing dual intelligent boundary detection on the three-dimensional image of the first perspective and the three-dimensional sub-image of the second perspective, avoiding the inaccurate boundary detection results caused by mutual occlusion between objects at a single perspective, improving the accuracy of the suspicious item boundary detection results, and then using the detected boundary of the suspicious item to mark the three-dimensional area of ​​the suspicious item in the three-dimensional image, improving the accuracy of the suspicious item marking results. In addition, the electronic device can perform intelligent detection on the boundary of the suspicious item by itself, and the user does not need to draw the boundary of the suspicious item multiple times, but only needs to confirm the boundary of the suspicious item detected by the electronic device to realize the marking of the suspicious item. The operation is simple and the marking process takes a short time, which improves the user's image reading efficiency and item marking efficiency, thereby improving the overall security inspection efficiency.

[0080] In public places, security inspection of the items carried by people is an important measure to ensure public safety. During the security inspection process, the items carried by people are first placed in the security inspection equipment as the inspected items, wherein the security inspection equipment can be a CT (Computed Tomography) device, a single-energy X-ray detection device, or a dual-energy X-ray detection device, etc., without limitation. When the inspected items are transmitted to the detection area, the X-rays emitted by the ray source in the security inspection equipment can pass through the inspected items and be detected by the detector to obtain the three-dimensional data of the inspected items. Then the electronic device processes the three-dimensional data of the inspected items to obtain a three-dimensional image of the inspected items. The staff in the security inspection area can read the three-dimensional image to view the detailed information inside the inspected items, and determine whether there are suspicious items inside the inspected items that need to be re-inspected. If it is determined that there are suspicious items inside, the suspicious items are marked and re-inspected to eliminate public safety hazards.

[0081] In order to improve the efficiency of marking suspicious objects, in the above step S101, the electronic device can obtain a three-dimensional image of the object to be inspected and display the three-dimensional image.

[0082] In order to mark suspicious items in the inspected object, the electronic device may first obtain a three-dimensional image of the inspected object and display the three-dimensional image of the inspected object in a preset display area. Figure 2 As shown, the three-dimensional image displays various items included in the inspected object, and the three-dimensional image is a rotatable three-dimensional image. The user can rotate the three-dimensional image to any angle in the o-xyz coordinate system to observe the three-dimensional image from the rotated viewing angle to determine whether there are suspicious items inside the inspected object, and when it is found that there are suspicious items inside the inspected object, the three-dimensional image can be continued to be rotated until the three-dimensional image is rotated to a viewing angle suitable for observing the suspicious items. The user can trigger the suspicious item marking function of the electronic device to instruct the electronic device to mark the suspicious items in the three-dimensional image. For example, the user can select the suspicious item marking function option in the display interface to trigger the suspicious item marking function of the electronic device, or the user can stay at the angle for a preset trigger time to trigger the suspicious item marking function of the electronic device, etc.

[0083] In one implementation, the electronic device can obtain three-dimensional data of the object under test, and reconstruct and render the three-dimensional data to display a three-dimensional image of the object under test in a preset display area. The three-dimensional data can be obtained by performing CT scanning, dual-energy X-ray detection, and other detection methods on the object under test, which are not limited here.

[0084] In one implementation, the electronic device can directly obtain a three-dimensional image of the object to be inspected, and display the obtained three-dimensional image on a preset display interface. For example, the electronic device can receive a three-dimensional image sent by another electronic device used to generate a three-dimensional image of the object to be inspected, and obtain the three-dimensional image of the object to be inspected; the electronic device can respond to the image identifier carried by the suspicious object marking instruction issued by the user, obtain the three-dimensional image with the image identifier from the three-dimensional image database, and obtain the three-dimensional image of the object to be inspected, etc.

[0085] Furthermore, in the above step S102, the first boundary of each object included in the inspected object in the three-dimensional image of the first viewing angle is detected, and the detected first boundary is displayed.

[0086] When displaying the three-dimensional image of the object under inspection at the first viewing angle, the electronic device can detect the first boundary of each item included in the object under inspection in the three-dimensional image of the first viewing angle, and display the detected first boundary of each item. The first viewing angle can be any viewing angle.

[0087] In one implementation, during the process of displaying a three-dimensional image by an electronic device, a user may rotate the three-dimensional image to adjust the viewing angle of the three-dimensional image. When the three-dimensional image is rotated to a first viewing angle that is convenient for viewing a suspicious object, the user may perform a boundary detection operation on the electronic device to trigger a suspicious object marking function. When the electronic device detects the boundary detection operation issued by the user, it may respond to the boundary detection operation to detect the first boundary of each object included in the object to be inspected in the three-dimensional image of the first viewing angle, and display the first boundary of each detected object.

[0088] In one implementation, the electronic device may use the viewing angle of the currently displayed three-dimensional image as the first viewing angle, and detect the first boundary of each object included in the inspected object in the three-dimensional image of the first viewing angle.

[0089] The electronic device can directly perform boundary detection on the three-dimensional image of the first perspective, or it can obtain a first target two-dimensional image of the first perspective corresponding to the three-dimensional image of the first perspective, and perform boundary detection on the first target two-dimensional image of the first perspective to obtain the first boundaries of each object included in the inspected object in the three-dimensional image of the first perspective.

[0090] Furthermore, the electronic device may detect the first boundaries of each object included in the three-dimensional image of the first perspective by using methods such as image intelligent detection or image segmentation to obtain the first boundaries of each object included in the three-dimensional image of the first perspective.

[0091] For example, Figure 2The electronic device can rotate the three-dimensional image to the first viewing angle based on the user's rotation operation on the three-dimensional image, and display the three-dimensional image in the preset display interface as shown in FIG. Figure 3 The user can select the suspicious object marking function option in the display interface to perform a boundary detection operation on the three-dimensional image of the first perspective. The electronic device can respond to the boundary detection operation performed by the user on the three-dimensional image of the first perspective, obtain a first target two-dimensional image of the first perspective corresponding to the three-dimensional image of the first perspective, and use a two-dimensional image segmentation algorithm to perform boundary detection on the first target two-dimensional image of the first perspective, and obtain the following Figure 4 The first boundaries of the objects 1, 2, 3 and 4 included in the three-dimensional image of the first viewing angle are shown, and the first boundaries of the objects 1-4 in the three-dimensional image of the first viewing angle are also shown.

[0092] According to the first boundaries of each item included in the inspected object displayed by the electronic device, the user can perform a confirmation operation on the first target boundary of the suspicious item, and in the above step S103, the electronic device can respond to the user's confirmation operation on the first target boundary of the suspicious item, and determine in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious item indicated by the first target boundary at the second viewing angle. The second viewing angle is different from the first viewing angle.

[0093] Since the first boundary is obtained by the electronic device performing boundary detection on the three-dimensional image of the first perspective, the detected first boundary may be the boundary of each object with the smallest depth value in the depth direction of the first perspective. When there are other objects in the depth direction, the other objects may be partially or even completely blocked by the object with the smallest depth value and cannot be detected. Correspondingly, the boundaries of other objects may also be partially or even completely blocked by the object with the smallest depth value and cannot be detected. Alternatively, at the first perspective, if an object with a smaller depth value and a smaller volume overlaps an object with a larger depth value and a larger volume, then when performing boundary detection, the electronic device may mistakenly identify the smaller volume object and the larger volume object as one object, and then only detect the boundary of the larger volume object, but cannot detect the boundary of the smaller volume object. Since at least one of the above situations may exist, each first boundary detected by the electronic device may be the first boundary of multiple objects in the depth direction corresponding to the first boundary, and the suspicious object corresponding to the first boundary cannot be accurately determined based on the first boundary, and further, the three-dimensional area where the suspicious object is located cannot be accurately determined based on the first boundary.

[0094] Based on this, when the user performs a confirmation operation on the first target boundary of the suspicious object according to the detected first boundaries, in order to accurately determine the three-dimensional area where the suspicious object is located, the electronic device can respond to the confirmation operation and determine the three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary in the three-dimensional image at the second viewing angle. The second viewing angle is different from the first viewing angle, and there is an angle conversion relationship between the second viewing angle and the first viewing angle, that is, there is an angle difference between the second viewing angle and the first viewing angle, and the angle difference value is not 0. For example, the angle difference between the first viewing angle and the second viewing angle is 90 degrees.

[0095] The second viewing angle may be a viewing angle selected by a user, or may be a second viewing angle calculated by the electronic device based on the first viewing angle and a preset angle difference, both of which are reasonable and are not specifically limited here.

[0096] That is to say, based on the first target boundary, the electronic device can determine in the three-dimensional image a three-dimensional sub-graph corresponding to the first target boundary in the depth direction of the first perspective, and since observing the three-dimensional sub-graph from the first perspective cannot solve the problem of mutual occlusion of objects in the depth direction of the first perspective, in order to accurately determine the three-dimensional area of ​​the suspicious object, the three-dimensional sub-graph can be observed from a second perspective different from the first perspective, that is, the three-dimensional sub-graph of the first perspective is rotated from the first perspective to the second perspective to obtain the three-dimensional sub-graph of the second perspective corresponding to the first target boundary.

[0097] According to the user's determination operation of the first target boundary in the three-dimensional image of the suspicious object at the first perspective, the three-dimensional sub-image corresponding to the suspicious object at the second perspective is determined, which can narrow the detection range of the secondary boundary detection. In this way, the three-dimensional area of ​​the suspicious object is determined by the three-dimensional sub-image of a part of the displayed three-dimensional image at another perspective, which can improve the efficiency and accuracy of the secondary boundary detection.

[0098] Furthermore, in the above step S104, the electronic device may detect the second boundary of each object in the three-dimensional sub-image of the second viewing angle, and determine the second target boundary of the suspicious object based on the detected second boundary.

[0099] After determining the three-dimensional sub-image of the second viewing angle, the electronic device may continue to detect the second boundary of each object in the three-dimensional sub-image of the second viewing angle.

[0100] Among them, the electronic device can directly perform boundary detection on the three-dimensional sub-image of the second perspective, or obtain a second target two-dimensional image of the second perspective corresponding to the three-dimensional sub-image of the second perspective, and perform boundary detection on the second target two-dimensional image to obtain the second boundaries of each object in the three-dimensional sub-image of the second perspective.

[0101] Furthermore, the electronic device may detect the second boundary of each object in the three-dimensional sub-image of the second perspective by using methods such as image intelligent detection or image segmentation to obtain the second boundary of each object in the three-dimensional sub-image of the second perspective.

[0102] After detecting the second boundaries of the objects in the three-dimensional sub-image of the second viewing angle, the electronic device may determine the second target boundaries of the suspicious object based on the detected second boundaries.

[0103] In one implementation, when there are multiple detected second boundaries, the user can select a target second boundary of the suspicious item from each detected second boundary. When the electronic device detects the user's selection operation on the second boundary, it can use the second boundary indicated by the selection operation as the second target boundary of the suspicious item.

[0104] In one implementation, when the number of detected second boundaries is multiple, the detected second boundaries do not include the second boundary of the suspicious object, or the detected second boundary of the suspicious object is incomplete, the user can draw the second target boundary of the suspicious object in a preset display interface, and when the electronic device detects the second target boundary drawn by the user, it can display the second target boundary drawn by the user in the preset display interface.

[0105] In one implementation, when the number of the detected second boundaries is 1, the electronic device may use the detected second boundary as the second target boundary of the suspicious object.

[0106] S105: Based on the second target boundary and the three-dimensional sub-image, determine a three-dimensional region of the suspicious object in the three-dimensional image, and mark it as the three-dimensional region of the suspicious object.

[0107] Based on the second target boundary of the suspicious object, the electronic device can determine the three-dimensional area corresponding to the second target boundary in the three-dimensional sub-image. Since the three-dimensional sub-image of the second perspective is determined based on the three-dimensional image of the first perspective, then according to the conversion relationship between the three-dimensional sub-image and the three-dimensional image, the area corresponding to the second target boundary of the suspicious object in the three-dimensional sub-image can be converted to the three-dimensional image to obtain the three-dimensional area where the suspicious object is located in the three-dimensional image of the first perspective, and the three-dimensional area is marked, that is, the three-dimensional area where the suspicious object is located is marked in the three-dimensional image. For example, Figure 5 As shown, the area 510 in the figure is the three-dimensional area where the suspicious items included in the inspected object marked in the three-dimensional image are located.

[0108] In the technical solution provided by the embodiment of the present application, the electronic device can determine the boundaries of the suspicious object at two different perspectives by performing dual intelligent boundary detection on the three-dimensional image of the first perspective and the three-dimensional sub-image of the second perspective, thereby avoiding the inaccurate boundary detection result caused by mutual occlusion between objects at a single perspective, improving the accuracy of the suspicious object boundary detection result, and then using the detected boundary of the suspicious object to mark the three-dimensional area of ​​the suspicious object in the three-dimensional image, thereby improving the accuracy of the suspicious object marking result. In addition, the electronic device can perform intelligent detection on the boundary of the suspicious object by itself, and the user does not need to draw the boundary of the suspicious object multiple times, but only needs to confirm the boundary of the suspicious object detected by the electronic device to realize the marking of the suspicious object. The operation is simple and the marking process takes a short time, which improves the user's image reading efficiency and item marking efficiency, thereby improving the overall security inspection efficiency.

[0109] As an implementation of an embodiment of the present application, in the above step S102, that is, the step of detecting the first boundary of each object included in the inspected object in the three-dimensional image of the first viewing angle and displaying the detected first boundary may include:

[0110] Acquire a first target two-dimensional image of the first perspective corresponding to the three-dimensional image, and perform boundary detection on the first target two-dimensional image to obtain first boundaries of each item included in the inspected object; display the detected first boundaries in the first target two-dimensional image; or, based on a conversion relationship between the first target two-dimensional image and the three-dimensional image, draw the detected first boundaries into the three-dimensional image, and display the first boundaries in the three-dimensional image.

[0111] Since intelligent detection or image segmentation methods are used to detect the boundaries of three-dimensional images to obtain the boundaries of each object in the three-dimensional image, more hardware resources are consumed, the requirements for hardware equipment are high, the boundary detection is time-consuming, and the boundary detection efficiency is low.

[0112] In order to save hardware resources and improve the efficiency of boundary detection, a first target two-dimensional image of the first viewing angle corresponding to the three-dimensional image of the first viewing angle may be acquired.

[0113] Among them, the electronic device can obtain the first target two-dimensional image of the first perspective corresponding to the three-dimensional image by taking a screenshot of a preset display interface displaying the three-dimensional image of the first perspective, or projecting the three-dimensional image of the first perspective in the depth direction of the first perspective.

[0114] Afterwards, the electronic device can perform boundary detection on the first target two-dimensional image to obtain the first boundaries of each item included in the inspected object. In order to facilitate the user to determine the first target boundary of the suspicious item based on the detected first boundaries, after determining the first boundary, the electronic device can display the first boundaries of each detected item in the first target two-dimensional image. Since the resources required for boundary detection of a two-dimensional image are less and the detection time is shorter, the electronic device performs boundary detection on the first target two-dimensional image, and while obtaining the first boundaries of each item included in the inspected object in the three-dimensional image of the first perspective, it reduces the equipment resources and detection time required for boundary detection, thereby improving the efficiency of boundary detection.

[0115] In order to allow users to intuitively view the boundary detection results in the three-dimensional image, after obtaining the first boundaries of each item included in the inspected object, the electronic device can also draw the first boundaries detected in the first target two-dimensional image into the three-dimensional image based on the conversion relationship between the first target two-dimensional image and the three-dimensional image, and display the detected first boundaries of each item included in the inspected object in the three-dimensional image, so that the user can more intuitively understand the spatial position of the first boundaries of each item included in the inspected object in the three-dimensional image.

[0116] For example, the electronic device may Figure 3 The three-dimensional image of the first perspective is projected to obtain a first target two-dimensional image of the first perspective, and the first boundaries of each object included in the object to be inspected in the first target two-dimensional image are detected, and the detected first boundaries are displayed in the first target two-dimensional image to obtain the following Figure 4 The first target two-dimensional image shown shows a first boundary between item 1 and item 4 .

[0117] In this embodiment, the electronic device can obtain a first target two-dimensional image of a first perspective corresponding to the three-dimensional image, and perform boundary detection on the first target two-dimensional image to obtain the first boundary of each item included in the inspected object; display the detected first boundary in the first target two-dimensional image; or based on the conversion relationship between the first target two-dimensional image and the three-dimensional image, draw the detected first boundary into the three-dimensional image, and display the first boundary in the three-dimensional image. In this way, by performing boundary detection on the first target two-dimensional image and displaying the first boundary of each item in the first target two-dimensional image or the three-dimensional image, the first boundary of each item included in the inspected object can be obtained, and the equipment resources required for the boundary detection of the item can be reduced, thereby improving the efficiency of boundary detection.

[0118] As an implementation of an embodiment of the present application, in the above step S102, that is, the step of detecting the first boundary of each object included in the inspected object in the three-dimensional image of the first viewing angle and displaying the detected first boundary may include:

[0119] Boundary detection is performed on the three-dimensional image to obtain first boundaries of each item included in the inspected object, and the detected first boundaries are displayed in the three-dimensional image.

[0120] The electronic device can directly perform boundary detection on the three-dimensional image to obtain the first boundary of each item included in the inspected object in the three-dimensional image. In addition, in order to facilitate the user to intuitively understand the first boundary of each item and determine the first target boundary of the suspicious item based on the detected first boundaries, the electronic device can display the first boundary of each detected item in the three-dimensional image. In this way, by directly performing boundary detection on the three-dimensional image and displaying the first boundary of each detected item in the three-dimensional image, it is convenient for the user to more intuitively understand the position of the first boundary of each item included in the inspected object in the three-dimensional image through the three-dimensional image of the first perspective, and then directly determine the first target boundary of the suspicious item in the three-dimensional image.

[0121] In this embodiment, the electronic device can perform boundary detection on the three-dimensional image, obtain the first boundary of each item included in the detected object, and display the detected first boundary in the three-dimensional image. In this way, by performing boundary detection on the three-dimensional image and displaying the first boundary of each item in the three-dimensional image, the user can intuitively understand the position of the boundary of each item in the three-dimensional image and perform subsequent first target boundary determination operations for suspicious items.

[0122] As an implementation of an embodiment of the present application, the above step S103, i.e., in response to the user's confirmation operation on the first target boundary of the suspicious object, determining, in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at the second viewing angle, may include:

[0123] In the case where the first boundary is displayed in the first target two-dimensional image, in response to a selection operation of the user on the first target boundary, or in response to a drawing operation of the user on the first target boundary, based on a conversion relationship between the first target two-dimensional image and the three-dimensional image, the first target boundary is drawn into the three-dimensional image; in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at the second viewing angle is intercepted;

[0124] When the first boundary is displayed in the three-dimensional image, in response to the user's selection operation on the first target boundary, or in response to the user's drawing operation on the first target boundary, in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at the second perspective is captured.

[0125] The user can determine the first target boundary of the suspicious object in the detected object based on the first boundary of each object included in the detected object in the three-dimensional image of the first perspective displayed in the preset display interface, and perform a confirmation operation on the first target boundary of the suspicious object. The confirmation operation can be a selection operation or a drawing operation.

[0126] Specifically, when the detected first boundary includes the first target boundary of the suspicious object, the user may perform a selection operation on the first target boundary. For example, the user may perform a selection operation on the first target boundary through an input device such as a mouse.

[0127] If the detected first boundary does not include the first target boundary of the suspicious object or the detected first target boundary of the suspicious object is incomplete, the user can perform a drawing operation for the first target boundary. For example, the user can draw the first target boundary of the suspicious object in the image showing the detected first boundaries of each object through the input device.

[0128] Accordingly, when the electronic device detects a confirmation operation performed by the user on a first target boundary of a suspicious object, it can cut out a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary in the three-dimensional image at the first perspective, based on the first target boundary determined by the user, and rotate the three-dimensional sub-image to the second perspective to obtain a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at the second perspective.

[0129] Since the electronic device can use the first target two-dimensional image of the first perspective corresponding to the three-dimensional image to perform boundary detection and display the first boundary of each detected object in the first target two-dimensional image or the three-dimensional image, or can use the three-dimensional image to perform boundary detection and display the first boundary of each detected object in the three-dimensional image, thus according to the image used by the electronic device for boundary detection, the image used to display the first boundary of each detected object later, and the specific situation of each detected first boundary, the confirmation operation performed by the user for the first target boundary can be different, and the image indicated by the confirmation operation of the first target boundary performed by the user can also be different. Accordingly, the subsequent electronic device performs the three-dimensional sub-image determination step in different ways, which can be specifically divided into the following situations:

[0130] The first case: the electronic device can perform boundary detection on the first target two-dimensional image of the first viewing angle corresponding to the three-dimensional image, and display the first boundary of each detected object in the first target two-dimensional image. In the case where the detected first boundary includes the first target boundary of the suspicious object, the user can perform a selection operation on the first target boundary. In response to the user's selection operation on the first target boundary, the electronic device can draw the first target boundary into the three-dimensional image based on the conversion relationship between the first target two-dimensional image and the three-dimensional image, and then intercept the three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary in the second viewing angle in the three-dimensional image.

[0131] The second case: the electronic device can perform boundary detection on the first target two-dimensional image of the first perspective corresponding to the three-dimensional image, and display the first boundary of each detected object in the first target two-dimensional image. In the case where the detected first boundary does not include the first target boundary of the suspicious object or the detected first target boundary of the suspicious object is incomplete, the user can perform a drawing operation on the first target boundary. In response to the user's drawing operation on the first target boundary, the electronic device can draw the first target boundary into the three-dimensional image based on the conversion relationship between the first target two-dimensional image and the three-dimensional image, and then intercept the three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary in the three-dimensional image at the second perspective.

[0132] The third case: the electronic device can perform boundary detection on the first target two-dimensional image of the first perspective corresponding to the three-dimensional image, and based on the conversion relationship between the first target two-dimensional image and the three-dimensional image, draw the detected first boundary into the three-dimensional image, and display the first boundary in the three-dimensional image. In the case where the detected first boundary includes the first target boundary of the suspicious object, the user can perform a selection operation on the first target boundary. In response to the user's selection operation on the first target boundary, the electronic device can intercept the three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary in the second perspective in the three-dimensional image.

[0133] Fourth situation: The electronic device can perform boundary detection on the first target two-dimensional image of the first perspective corresponding to the three-dimensional image, and based on the conversion relationship between the first target two-dimensional image and the three-dimensional image, draw the detected first boundary into the three-dimensional image, and display the first boundary in the three-dimensional image. In the case where the detected first boundary does not include the first target boundary of the suspicious object or the detected first target boundary of the suspicious object is incomplete, the user can perform a drawing operation on the first target boundary. In response to the user's drawing operation on the first target boundary, the electronic device can intercept the three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary in the second perspective in the three-dimensional image.

[0134] Fifth case: The electronic device may perform boundary detection on the three-dimensional image and display the detected first boundary in the three-dimensional image. In the case where the detected first boundary includes the first target boundary of the suspicious object, the user may perform a selection operation on the first target boundary. In response to the user's selection operation on the first target boundary, the electronic device may intercept the three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary in the three-dimensional image at the second viewing angle.

[0135] The sixth case: the electronic device can perform boundary detection on the three-dimensional image and display the detected first boundary in the three-dimensional image. In the case where the detected first boundary does not include the first target boundary of the suspicious object or the detected first target boundary of the suspicious object is incomplete, the user can perform a drawing operation on the first target boundary. In response to the user's drawing operation on the first target boundary, the electronic device can intercept the three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary in the three-dimensional image at the second viewing angle.

[0136] For example, the user Figure 4 In response to the user's selection operation, the electronic device can intercept the three-dimensional sub-image corresponding to the second viewing angle of the suspicious object indicated by the first target boundary in the three-dimensional image to obtain the following Figure 6 or Figure 7 The three-dimensional subgraph shown in Figure 2. Figure 6 The three-dimensional sub-image shown includes only one suspicious object, while Figure 7 The three-dimensional sub-image shown includes two items.

[0137] In this embodiment, when the first boundary is displayed in the first target two-dimensional image, the electronic device can respond to the user's selection operation or drawing operation on the first target boundary, based on the conversion relationship between the first target two-dimensional image and the three-dimensional image, draw the first target boundary into the three-dimensional image, and intercept the three-dimensional sub-image corresponding to the second perspective of the suspicious object indicated by the first target boundary in the three-dimensional image. In the case of displaying the first boundary in the three-dimensional image, the electronic device can respond to the user's drawing operation or drawing operation on the first target boundary, and intercept the three-dimensional sub-image corresponding to the second perspective of the suspicious object indicated by the first target boundary in the three-dimensional image. In this way, according to the image displaying each first boundary and the specific situation of the first boundary, the user can perform a corresponding confirmation operation to confirm the first target boundary of the suspicious object, and the electronic device can also perform steps corresponding to the user's confirmation operation to intercept the three-dimensional sub-image corresponding to the second perspective of the suspicious object indicated by the first perspective in the three-dimensional image, thereby improving the user experience and increasing the scope of application of the object marking method.

[0138] As an implementation of the embodiment of the present application, the above step S104, i.e., the step of detecting the second boundary of each object in the three-dimensional sub-image of the second viewing angle, may include:

[0139] A second target two-dimensional image of the second viewing angle corresponding to the three-dimensional sub-image is acquired, and boundary detection is performed on the second target two-dimensional image to obtain a second boundary of each object in the second target two-dimensional image.

[0140] Since using intelligent detection or image segmentation methods to detect the boundaries of objects in a three-dimensional image will consume more hardware resources, have higher requirements on hardware devices, take a long time to detect, and have low efficiency in boundary detection.

[0141] In order to save hardware resources and improve the efficiency of boundary detection, a second target two-dimensional image of the second perspective corresponding to the three-dimensional sub-image of the second perspective may be obtained.

[0142] Among them, the electronic device can obtain the second target two-dimensional image of the second perspective corresponding to the three-dimensional sub-image by taking a screenshot of a preset display interface displaying the three-dimensional sub-image of the second perspective, or by projecting the three-dimensional sub-image of the second perspective in the depth direction of the second perspective.

[0143] Afterwards, the electronic device can perform boundary detection on the second target two-dimensional image to obtain the second boundary of each object in the second target two-dimensional image. In this way, by performing boundary detection on the second target two-dimensional image, while obtaining the second boundary of each object included in the inspected object, the equipment resources required for boundary detection are reduced, the detection time is shortened, and the boundary detection efficiency is improved.

[0144] For example, the electronic device may Figure 6 The three-dimensional sub-image of the second perspective shown in FIG. 1 is projected to obtain a second target two-dimensional image of the second perspective, and then the second boundary of each object in the second target two-dimensional image is detected, and each detected second boundary is displayed in the second target two-dimensional image, so as to obtain the following Figure 8 The second target two-dimensional image shown shows a second boundary of the object 1 .

[0145] In this embodiment, the electronic device can obtain the second target two-dimensional image of the second perspective corresponding to the three-dimensional sub-image, and perform boundary detection on the second target two-dimensional image to obtain the second boundary of each object in the second target two-dimensional image. In this way, by performing boundary detection on the second target two-dimensional image, the second boundary of each object in the second target two-dimensional image can be obtained, and the equipment resources required for object boundary detection can be reduced, the boundary detection time can be shortened, and the boundary detection efficiency can be improved.

[0146] As an implementation of the embodiment of the present application, the above step S104, i.e., the step of detecting the second boundary of each object in the three-dimensional sub-image of the second viewing angle, may include:

[0147] Boundary detection is performed on the three-dimensional sub-image to obtain a second boundary of each object in the three-dimensional sub-image.

[0148] After acquiring the 3D sub-image of the second viewing angle, the electronic device can directly perform boundary detection on the 3D sub-image to obtain the second boundary of each object in the 3D sub-image. In this way, by directly performing boundary detection on the 3D sub-image, the position of the second boundary of each object in the 3D sub-image can be directly determined.

[0149] For example, the electronic device determines Figure 7 The three-dimensional sub-image of the second viewing angle shown in FIG. 1 is used to perform boundary detection, and each second boundary detected is displayed in the three-dimensional sub-image, so as to obtain the following Fig. 9 The three-dimensional sub-image shown shows the second boundary of item 1 and item 2.

[0150] In this embodiment, the electronic device can perform boundary detection on the three-dimensional sub-image to obtain the second boundary of each object in the three-dimensional sub-image, and can directly determine the position of the second boundary of each object in the three-dimensional sub-image. Subsequently, each second boundary is displayed in the three-dimensional sub-image, so that the user can intuitively understand the position of the boundary of each object in the three-dimensional sub-image, so that the user can perform the second target boundary determination operation of the suspicious object.

[0151] As an implementation of an embodiment of the present application, the above step S104, i.e., the step of determining the second target boundary of the suspicious object based on the detected second boundary, may include:

[0152] When the number of the detected second boundaries is 1, determining the detected second boundaries as the second target boundaries of the suspicious object;

[0153] In the case that the number of detected second boundaries is multiple, the second boundary indicated by the user's selection operation on the second boundary, or the boundary drawn by the user, is determined as the second target boundary of the suspicious object.

[0154] Since the three-dimensional sub-image of the second perspective is determined based on the first target boundary of the suspicious object, the three-dimensional sub-image is actually a three-dimensional image corresponding to the first target boundary in the depth direction of the first perspective in the three-dimensional image, and is a part of the three-dimensional image of the inspected object. The three-dimensional sub-image may include multiple objects or only one object.

[0155] Based on this, after the electronic device performs boundary detection on the three-dimensional sub-image of the second viewing angle, the number of second boundaries of each object obtained may be one or more.

[0156] like Figure 8 As shown, when the number of detected second boundaries is 1, the electronic device can directly determine the detected second boundary as the second target boundary of the suspicious object 1.

[0157] And as Fig. 9 As shown, when the number of detected second boundaries is multiple, the user can determine the second target boundary of the suspicious object 1 based on the second boundaries of each object in the three-dimensional sub-image detected by the electronic device displayed in the preset display interface, and perform a confirmation operation for the second target boundary of the suspicious object.

[0158] Specifically, when the detected second boundary includes the second target boundary of the suspicious object, the user may perform a selection operation on the second boundary. For example, the user may perform a selection operation on the second boundary through an input device such as a mouse.

[0159] If the detected second boundary does not include the second target boundary of the suspicious object or the detected second target boundary of the suspicious object is incomplete, the user can draw the boundary of the suspicious object by himself. For example, the user can draw the boundary of the suspicious object in the image displayed by the preset display device through the input device.

[0160] Correspondingly, when the electronic device detects a selection operation performed by the user on the second boundary, it can determine the second boundary indicated by the selection operation as the second target boundary of the suspicious object; when the electronic device detects a drawing operation performed by the user, it can determine the boundary drawn by the user as the second target boundary of the suspicious object.

[0161] In one implementation, when the number of detected second boundaries is 1, if the detected second boundary is not the second target boundary of the suspicious object, or the detected second target boundary of the suspicious object is incomplete, the user can draw the second boundary of the suspicious object by himself. When the electronic device detects the user's drawing operation, it can determine the boundary drawn by the user as the second target boundary of the suspicious object.

[0162] In this embodiment, when the number of detected second boundaries is 1, the electronic device can determine the detected second boundary as the second target boundary of the suspicious object; and when the number of detected second boundaries is multiple, the second boundary indicated by the user's selection operation for the second boundary or the boundary drawn by the user is determined as the second target boundary of the suspicious object. In this way, according to the specific situation of the detected second boundary, the electronic device can perform corresponding operations to determine the second target boundary of the suspicious object, and by determining the two boundaries of the suspicious object at the first viewing angle and the second viewing angle respectively, the double detection of the boundary of the suspicious object is achieved, thereby improving the reliability of the detection result of the boundary of the suspicious object.

[0163] As an implementation of an embodiment of the present application, when the number of detected second boundaries is multiple, before the above step of determining the second boundary indicated by the user's selection operation on the second boundary, or the boundary drawn by the user, as the second target boundary of the suspicious object, an object marking method provided by an embodiment of the present application may further include:

[0164] In the case of performing boundary detection on the second target two-dimensional image, displaying the detected second boundary in the second target two-dimensional image, or, based on a conversion relationship between the second target two-dimensional image and the three-dimensional sub-image, drawing the detected second boundary into the three-dimensional sub-image, and displaying the second boundary in the three-dimensional sub-image;

[0165] When the boundary detection is performed on the three-dimensional sub-image, the detected second boundary is displayed in the three-dimensional sub-image.

[0166] In order to facilitate the user to determine the second target boundary of the suspicious object based on the detected second boundary, the electronic device can display the detected second boundary of each object in a preset display interface.

[0167] Since the electronic device can achieve the purpose of detecting the second boundary of each object in the three-dimensional sub-image by performing boundary detection on the second target two-dimensional image of the second perspective corresponding to the three-dimensional sub-image, or by directly performing boundary detection on the three-dimensional sub-image, under different boundary detection methods, the electronic device can also use different methods to display the second boundary of each detected object.

[0168] In one case, when performing boundary detection on the second target two-dimensional image, the electronic device may display the detected second boundary in the second target two-dimensional image.

[0169] In one case, when performing boundary detection on the second target two-dimensional image, the electronic device can draw the detected second boundary into the three-dimensional sub-image based on the conversion relationship between the second target two-dimensional image and the three-dimensional sub-image, and display the detected second boundary of each object in the three-dimensional sub-image.

[0170] In another case, when performing boundary detection on the three-dimensional sub-image, the electronic device may directly display the detected second boundary in the three-dimensional sub-image.

[0171] Based on this, according to the image used by the electronic device for boundary detection, the image subsequently used to display the second boundaries of each detected object, and the specific circumstances of each detected second boundary, the confirmation operation performed by the user for the second target boundary may be different, and the image indicated by the confirmation operation performed by the user may also be different. Accordingly, the method for determining the second target boundary of the suspicious object that the electronic device needs to execute may also be different.

[0172] In one case, when performing boundary detection on the second target two-dimensional image, the electronic device may display the detected second boundary in the second target two-dimensional image, and when the detected second boundary includes the second target boundary of the suspicious object, the user may perform a selection operation on the second boundary in the second target two-dimensional image. The electronic device may determine the second boundary indicated by the user's selection operation on the second boundary as the second target boundary of the suspicious object.

[0173] In one case, the electronic device may perform boundary detection on the second target two-dimensional image and display the detected second boundary in the second target two-dimensional image. When the detected first boundary does not include the second target boundary of the suspicious object or the detected second target boundary of the suspicious object is incomplete, the user may draw the second target boundary of the suspicious object in the second target two-dimensional image. The electronic device may determine the second boundary drawn by the user as the second target boundary of the suspicious object.

[0174] In one case, the electronic device may perform boundary detection on the second target two-dimensional image, and based on the conversion relationship between the second target two-dimensional image and the three-dimensional image, draw the detected second boundary into the three-dimensional sub-image, and display the second boundary in the three-dimensional sub-image. In the case where the detected second boundary includes the second target boundary of the suspicious object, the user may perform a selection operation on the second boundary in the three-dimensional sub-image. The electronic device may determine the second boundary indicated by the user's selection operation on the second boundary as the second target boundary of the suspicious object.

[0175] In one case, the electronic device may perform boundary detection on the second target two-dimensional image, and based on the conversion relationship between the second target two-dimensional image and the three-dimensional image, draw the detected second boundary into the three-dimensional sub-image, and display the second boundary in the three-dimensional sub-image. In the case where the detected second boundary does not include the second target boundary of the suspicious object or the detected second target boundary of the suspicious object is incomplete, the user may draw the second target boundary of the suspicious object in the three-dimensional sub-image. The electronic device may determine the second boundary drawn by the user as the second target boundary of the suspicious object.

[0176] In one case, the electronic device may perform boundary detection on the three-dimensional sub-image and display the detected second boundary in the three-dimensional sub-image. In the case where the detected second boundary includes the second target boundary of the suspicious object, the user may perform a selection operation on the second boundary in the three-dimensional sub-image. The electronic device may determine the second boundary indicated by the user's selection operation on the second boundary as the second target boundary of the suspicious object.

[0177] In one case, the electronic device may perform boundary detection on the three-dimensional sub-image and display the detected second boundary in the three-dimensional sub-image. In the case where the detected second boundary does not include the second target boundary of the suspicious object or the detected second target boundary of the suspicious object is incomplete, the user may draw the second target boundary of the suspicious object in the three-dimensional sub-image. The electronic device may determine the second boundary drawn by the user as the second target boundary of the suspicious object.

[0178] In this embodiment, when performing boundary detection on the second target two-dimensional image, the electronic device can display the detected second boundary in the second target two-dimensional image, or based on the conversion relationship between the second target two-dimensional image and the three-dimensional sub-image, draw the detected second boundary into the three-dimensional sub-image and display the second boundary in the three-dimensional sub-image; and when performing boundary detection on the three-dimensional sub-image, the electronic device can display the detected second boundary in the three-dimensional sub-image. In this way, according to the image showing each second boundary and the specific situation of the second boundary, the user can perform corresponding operations to confirm the second target boundary of the suspicious object, and the electronic device can also perform steps corresponding to the user operation to determine the second target boundary of the suspicious object, thereby improving the user experience and increasing the scope of application of the object marking method.

[0179] As an implementation of an embodiment of the present application, the above step S105, i.e., the step of determining the three-dimensional area of ​​the suspicious object in the three-dimensional image based on the second target boundary and the three-dimensional sub-image, may include:

[0180] According to the conversion relationship between the first viewing angle and the second viewing angle, a three-dimensional area corresponding to the area corresponding to the second object boundary in the three-dimensional sub-image in the three-dimensional image is determined.

[0181] Since the 3D sub-image of the second perspective is intercepted from the 3D image of the first perspective, the conversion relationship between the 3D image of the first perspective and the 3D sub-image of the second perspective is the conversion relationship between the first perspective and the second perspective.

[0182] After determining the second target boundary of the suspicious object, the electronic device can determine the area corresponding to the second target boundary in the three-dimensional sub-image, and determine the three-dimensional area corresponding to the area corresponding to the second target boundary in the three-dimensional sub-image in the three-dimensional image according to the conversion relationship between the first perspective and the second perspective, to obtain the three-dimensional area of ​​the suspicious object in the three-dimensional image. For example, Figure 5 As shown, a three-dimensional region 510 of the suspicious object in the three-dimensional image can be represented by a bounding box.

[0183] In one implementation, the electronic device can determine the four vertex coordinates of the second target boundary in the three-dimensional sub-image, and then determine the conversion of the four vertex coordinates of the second target boundary in the three-dimensional sub-image into vertex coordinates in the three-dimensional image under the first perspective based on the conversion relationship between the first perspective and the second perspective, determine the regional position of the three-dimensional sub-image in the three-dimensional image based on the four vertex coordinates of the first target boundary, and then determine the regional position of the three-dimensional sub-image in the three-dimensional image and the vertex coordinates of the second target boundary in the three-dimensional image. The region corresponding to the second target boundary in the three-dimensional sub-image and the region corresponding to the second target boundary in the three-dimensional image can be determined based on the regional position of the three-dimensional sub-image in the three-dimensional image and the vertex coordinates of the second target boundary in the three-dimensional image, and the three-dimensional region where the suspicious object is located can be determined.

[0184] In one implementation, the electronic device can rotate the three-dimensional sub-image to the first perspective based on the conversion relationship between the first perspective and the second perspective, and determine the area corresponding to the second target boundary in the three-dimensional sub-image and the area corresponding to the three-dimensional sub-image in the three-dimensional image based on the area corresponding to the second target boundary in the three-dimensional sub-image at the first perspective, thereby determining the three-dimensional area in the three-dimensional image where the suspicious object is located.

[0185] In this embodiment, the electronic device can determine the three-dimensional area in the three-dimensional image corresponding to the area corresponding to the second target boundary in the three-dimensional sub-image according to the conversion relationship between the first perspective and the second perspective. In this way, by determining the first target boundary and the second target boundary of the suspicious object, the three-dimensional area where the suspicious object is located can be determined in the three-dimensional image, thereby achieving the purpose of marking the location of the suspicious object in the three-dimensional image. In addition, since the electronic device automatically draws the boundary of the object for the user to select by performing object boundary detection. In many application scenarios, especially in simple application scenarios such as airport baggage inspection, security inspectors usually do not need or only need to perform one boundary selection to complete the marking of suspicious objects, which greatly improves the efficiency of security inspectors in reading images, thereby improving security inspection efficiency.

[0186] As an implementation of the embodiment of the present application, the method further includes:

[0187] Obtaining the item category of the suspicious item;

[0188] The article category of the suspicious article and the three-dimensional area of ​​the suspicious article are recorded to obtain the marking result of the inspected article so that the staff can re-inspect the suspicious article.

[0189] After marking the three-dimensional area where the suspicious object is located in the three-dimensional image, the electronic device can also obtain the object category of the suspicious object.

[0190] For example, after marking the three-dimensional area of ​​the suspicious item, the user can enter the item category of the suspicious item in the item category input area of ​​the preset display interface, or the user can select the item category of the suspicious item from the preset item categories, and the electronic device can obtain the item category of the suspicious item input or selected by the user to obtain the item category of the suspicious item.

[0191] Afterwards, the electronic device can record the item category of the suspicious item and the three-dimensional area of ​​the suspicious item marked in the three-dimensional image, and obtain the marking result of the inspected object, so that the staff can subsequently re-inspect the suspicious item based on the marking result of the inspected object, that is, verify the item category to which the suspicious item belongs in the three-dimensional area of ​​the above-mentioned suspicious item. If the suspicious item belongs to the above-mentioned item category, the security inspection result of the inspected object is obtained.

[0192] In one implementation, the electronic device can record the coordinates of the first target boundary and the second target boundary of the suspicious object in the three-dimensional image and the item category of the suspicious object, and obtain the marking result of the inspected object for the staff to re-inspect the suspicious object in subsequent unpacking and re-inspection processes.

[0193] In this embodiment, the electronic device can obtain the category of the suspicious item, record the category of the suspicious item and the three-dimensional area of ​​the suspicious item, and obtain the marking result of the inspected item, so that the staff can re-inspect the suspicious item. In this way, the three-dimensional area of ​​the suspicious item is marked through double boundary detection, and the category of the suspicious item is marked, which improves the accuracy of the item marking, thereby greatly facilitating the subsequent re-inspection process and improving the overall security inspection efficiency.

[0194] Corresponding to the object marking method, the present embodiment also provides an object marking device.

[0195] like Fig.10 As shown, an article marking device comprises:

[0196] An image display module 1001 is used to obtain a three-dimensional image of the object under inspection and display the three-dimensional image;

[0197] A first boundary detection module 1002, configured to detect a first boundary of each object included in the detected object in the three-dimensional image of the first viewing angle, and display the detected first boundary;

[0198] The three-dimensional sub-image determination module 1003 is used to determine, in response to a user's confirmation operation on a first target boundary of a suspicious object, a three-dimensional sub-image corresponding to a second viewing angle of the suspicious object indicated by the first target boundary in the three-dimensional image; wherein the second viewing angle is different from the first viewing angle;

[0199] A second boundary detection module 1004, configured to detect a second boundary of each object in the three-dimensional sub-image of the second viewing angle, and determine a second target boundary of the suspicious object based on the detected second boundary;

[0200] The three-dimensional region determination module 1005 is used to determine the three-dimensional region of the suspicious object in the three-dimensional image based on the second target boundary and the three-dimensional sub-image, and mark it as the three-dimensional region of the suspicious object.

[0201] In the technical solution provided by the embodiment of the present application, the electronic device can obtain a three-dimensional image of the inspected object and display the three-dimensional image; detect the first boundary of each item included in the inspected object in the three-dimensional image of the first perspective, and display the detected first boundary; in response to the user's confirmation operation on the first target boundary of the suspicious item, determine in the three-dimensional image a three-dimensional sub-graph corresponding to the suspicious item indicated by the first target boundary at the second perspective; wherein the second perspective is different from the first perspective; detect the second boundary of each item in the three-dimensional sub-graph of the second perspective, and determine the second target boundary of the suspicious item based on the detected second boundary; based on the second target boundary and the three-dimensional sub-graph, determine the three-dimensional area of ​​the suspicious item in the three-dimensional image, and mark it as the three-dimensional area of ​​the suspicious item.

[0202] In the technical solution provided by the present application, the electronic device can determine the boundaries of suspicious items at two different perspectives by performing dual intelligent boundary detection on the three-dimensional image of the first perspective and the three-dimensional sub-image of the second perspective, thereby avoiding the inaccurate boundary detection results caused by mutual occlusion between items at a single perspective, improving the accuracy of the suspicious item boundary detection results, and then using the detected boundary of the suspicious item to mark the three-dimensional area of ​​the suspicious item in the three-dimensional image, thereby improving the accuracy of the suspicious item marking results. In addition, the electronic device can perform intelligent detection on the boundary of the suspicious item by itself, and the user does not need to draw the boundary of the suspicious item multiple times, but only needs to confirm the boundary of the suspicious item detected by the electronic device to realize the marking of the suspicious item. The operation is simple and the marking process takes a short time, which improves the user's image reading efficiency and item marking efficiency, thereby improving the overall security inspection efficiency.

[0203] As an implementation of the embodiment of the present application, the first boundary detection module 1002 includes:

[0204] A first detection unit is used to obtain a first target two-dimensional image of the first viewing angle corresponding to the three-dimensional image, and perform boundary detection on the first target two-dimensional image to obtain a first boundary of each item included in the detected object; display the detected first boundary in the first target two-dimensional image; or, based on a conversion relationship between the first target two-dimensional image and the three-dimensional image, draw the detected first boundary into the three-dimensional image, and display the first boundary in the three-dimensional image;

[0205] or,

[0206] The second detection unit is used to perform boundary detection on the three-dimensional image to obtain a first boundary of each object included in the detected object, and display the detected first boundary in the three-dimensional image.

[0207] As an implementation of the embodiment of the present application, the three-dimensional sub-image determination module 1003 includes:

[0208] A first sub-image determining unit is configured to, when the first boundary is displayed in the first target two-dimensional image, in response to a selection operation of the user on the first target boundary, or in response to a drawing operation of the user on the first target boundary, draw the first target boundary into the three-dimensional image based on a conversion relationship between the first target two-dimensional image and the three-dimensional image; and intercept, in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at a second viewing angle;

[0209] The second sub-image determining unit is used to intercept, in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at a second perspective in response to a selection operation of the user on the first target boundary or a drawing operation of the user on the first target boundary when the first boundary is displayed in the three-dimensional image.

[0210] As an implementation of the embodiment of the present application, the second boundary detection module 1004 includes:

[0211] A second boundary detection unit is used to obtain a second target two-dimensional image of the second perspective corresponding to the three-dimensional sub-image, and perform boundary detection on the second target two-dimensional image to obtain a second boundary of each object in the second target two-dimensional image; or, perform boundary detection on the three-dimensional sub-image to obtain a second boundary of each object in the three-dimensional sub-image.

[0212] As an implementation of the embodiment of the present application, the second boundary detection module 1004 includes:

[0213] a first determining unit, configured to determine the detected second boundary as a second target boundary of the suspicious object when the number of the detected second boundaries is 1;

[0214] The second determining unit is configured to determine, when a plurality of second boundaries are detected, a second boundary indicated by the user's selection operation on the second boundary, or a boundary drawn by the user, as a second target boundary of the suspicious object.

[0215] As an implementation of the embodiment of the present application, the device further includes:

[0216] A first display module is used for, before the step of determining the second boundary indicated by the user's selection operation on the second boundary, or the boundary drawn by the user, as the second target boundary of the suspicious object when the number of the detected second boundaries is multiple, displaying the detected second boundary in the second target two-dimensional image when boundary detection is performed on the second target two-dimensional image, or drawing the detected second boundary into the three-dimensional sub-image based on the conversion relationship between the second target two-dimensional image and the three-dimensional sub-image, and displaying the second boundary in the three-dimensional sub-image;

[0217] The second display module is used to display the detected second boundary in the three-dimensional sub-image when boundary detection is performed on the three-dimensional sub-image.

[0218] As an implementation of the embodiment of the present application, the three-dimensional area determination module 1005 includes:

[0219] The three-dimensional region determining unit is used to determine the three-dimensional region in the three-dimensional image corresponding to the region corresponding to the second target boundary in the three-dimensional sub-image according to the conversion relationship between the first viewing angle and the second viewing angle.

[0220] As an implementation of the embodiment of the present application, the device further includes:

[0221] A category acquisition module, used to acquire the category of the suspicious object;

[0222] The recording module is used to record the article category of the suspicious article and the three-dimensional area of ​​the suspicious article, and obtain the marking result of the inspected article so that the staff can re-inspect the suspicious article.

[0223] The present application also provides an electronic device, such as Fig.11 As shown, including:

[0224] Memory 1101, used for storing computer programs;

[0225] The processor 1102 is used to implement the object marking method provided in the embodiment of the present application when executing the program stored in the memory 1101.

[0226] In addition, the electronic device may further include a communication bus and / or a communication interface, and the processor 1102, the communication interface, and the memory 1101 communicate with each other via the communication bus.

[0227] In the technical solution provided by the embodiment of the present application, the electronic device can determine the boundaries of the suspicious object at two different perspectives by performing dual intelligent boundary detection on the three-dimensional image of the first perspective and the three-dimensional sub-image of the second perspective, thereby avoiding the inaccurate boundary detection result caused by mutual occlusion between objects at a single perspective, improving the accuracy of the suspicious object boundary detection result, and then using the detected boundary of the suspicious object to mark the three-dimensional area of ​​the suspicious object in the three-dimensional image, thereby improving the accuracy of the suspicious object marking result. In addition, the electronic device can perform intelligent detection on the boundary of the suspicious object by itself, and the user does not need to draw the boundary of the suspicious object multiple times, but only needs to confirm the boundary of the suspicious object detected by the electronic device to realize the marking of the suspicious object. The operation is simple and the marking process takes a short time, which improves the user's image reading efficiency and item marking efficiency, thereby improving the overall security inspection efficiency.

[0228] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0229] The communication interface is used for communication between the above electronic device and other devices.

[0230] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0231] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0232] In another embodiment provided in the present application, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned object marking methods are implemented.

[0233] In another embodiment provided by the present application, a computer program product including instructions is also provided, which, when executed on a computer, enables the computer to execute any one of the object marking methods in the above embodiments.

[0234] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part 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, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a solid-state hard disk (SSD), etc.

[0235] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0236] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, electronic device, computer-readable storage medium, and computer program product embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0237] The above description is only a preferred embodiment of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application are included in the protection scope of the present application.

Claims

1. A method for marking an object, characterized in that: The method comprises: Acquire a three-dimensional image of the object under inspection and display the three-dimensional image; Detecting a first boundary of each object included in the inspected object in the three-dimensional image of the first viewing angle, and displaying the detected first boundary; In response to a user's confirmation operation on a first target boundary of a suspicious object, determining, in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at a second viewing angle; wherein the second viewing angle is different from the first viewing angle; Detecting a second boundary of each object in the three-dimensional sub-image of the second viewing angle, and determining a second target boundary of the suspicious object based on the detected second boundary; Based on the second object boundary and the three-dimensional sub-image, a three-dimensional region of the suspicious object in the three-dimensional image is determined and marked as the three-dimensional region of the suspicious object.

2. The method according to claim 1, characterized in that The step of detecting a first boundary of each object included in the inspected object in the three-dimensional image of the first viewing angle and displaying the detected first boundary includes: Acquire a first target two-dimensional image of the first viewing angle corresponding to the three-dimensional image, and perform boundary detection on the first target two-dimensional image to obtain first boundaries of each item included in the inspected object; display the detected first boundaries in the first target two-dimensional image; or, based on a conversion relationship between the first target two-dimensional image and the three-dimensional image, draw the detected first boundaries into the three-dimensional image, and display the first boundaries in the three-dimensional image; or, Boundary detection is performed on the three-dimensional image to obtain first boundaries of each item included in the inspected object, and the detected first boundaries are displayed in the three-dimensional image.

3. The method according to claim 2, characterized in that In response to the user's confirmation operation on the first target boundary of the suspicious object, the step of determining, in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at the second viewing angle comprises: In the case where the first boundary is displayed in the first target two-dimensional image, in response to a selection operation of the user on the first target boundary, or in response to a drawing operation of the user on the first target boundary, based on a conversion relationship between the first target two-dimensional image and the three-dimensional image, the first target boundary is drawn into the three-dimensional image; in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at the second viewing angle is intercepted; When the first boundary is displayed in the three-dimensional image, in response to the user's selection operation on the first target boundary, or in response to the user's drawing operation on the first target boundary, in the three-dimensional image, a three-dimensional sub-image corresponding to the suspicious object indicated by the first target boundary at the second perspective is captured.

4. The method according to claim 1, characterized in that: The step of detecting a second boundary of each object in the three-dimensional sub-image of the second viewing angle comprises: Acquire a second target two-dimensional image of the second viewing angle corresponding to the three-dimensional sub-image, and perform boundary detection on the second target two-dimensional image to obtain a second boundary of each object in the second target two-dimensional image; or, Boundary detection is performed on the three-dimensional sub-image to obtain a second boundary of each object in the three-dimensional sub-image.

5. The method according to claim 4, characterized in that The step of determining the second target boundary of the suspicious object based on the detected second boundary includes: When the number of the detected second boundaries is 1, determining the detected second boundaries as the second target boundaries of the suspicious object; In the case that the number of detected second boundaries is multiple, the second boundary indicated by the user's selection operation on the second boundary, or the boundary drawn by the user, is determined as the second target boundary of the suspicious object.

6. The method according to claim 5, characterized in that In the case where the number of detected second boundaries is multiple, before the step of determining the second boundary indicated by the user's selection operation on the second boundary, or the boundary drawn by the user, as the second target boundary of the suspicious object, the method further includes: In the case of performing boundary detection on the second target two-dimensional image, displaying the detected second boundary in the second target two-dimensional image, or, based on a conversion relationship between the second target two-dimensional image and the three-dimensional sub-image, drawing the detected second boundary into the three-dimensional sub-image, and displaying the second boundary in the three-dimensional sub-image; When the boundary detection is performed on the three-dimensional sub-image, the detected second boundary is displayed in the three-dimensional sub-image.

7. The method according to any one of claims 1 to 6, characterized in that The step of determining the three-dimensional area of ​​the suspicious object in the three-dimensional image based on the second target boundary and the three-dimensional sub-image includes: According to the conversion relationship between the first viewing angle and the second viewing angle, a three-dimensional area corresponding to the area corresponding to the second object boundary in the three-dimensional sub-image in the three-dimensional image is determined.

8. The method according to any one of claims 1 to 6, characterized in that: The method further comprises: Obtaining the item category of the suspicious item; The article category of the suspicious article and the three-dimensional area of ​​the suspicious article are recorded to obtain the marking result of the inspected article so that the staff can re-inspect the suspicious article.

9. An object marking device, characterized in that: The device comprises: An image display module, used to obtain a three-dimensional image of the object under inspection and display the three-dimensional image; A first boundary detection module, used to detect a first boundary of each object included in the detected object in the three-dimensional image of the first viewing angle, and display the detected first boundary; a three-dimensional sub-image determination module, configured to determine, in response to a user's confirmation operation on a first target boundary of a suspicious object, a three-dimensional sub-image corresponding to a second viewing angle of the suspicious object indicated by the first target boundary in the three-dimensional image; wherein the second viewing angle is different from the first viewing angle; A second boundary detection module, configured to detect a second boundary of each object in the three-dimensional sub-image of the second viewing angle, and determine a second target boundary of the suspicious object based on the detected second boundary; A three-dimensional region determination module is used to determine the three-dimensional region of the suspicious object in the three-dimensional image based on the second target boundary and the three-dimensional sub-image, and mark it as the three-dimensional region of the suspicious object.

10. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor, for implementing any of the methods described in claims 1-8 when executing a program stored in a memory.

Images