Container truck anti-lifting detection method and related device

Abstract

The embodiments of this specification provide a container truck anti-lifting detection method and related devices, the method comprising: detecting the actual spatial attribute value of the designated target object included in the container truck from the reference image collected for the container truck; adaptively setting the target lifting judgment threshold for the container truck according to the conversion relationship between the actual spatial attribute value and the designated target value for the designated target object; based on the target lifting judgment threshold, performing anti-lifting detection on the collected on-site image of the container truck. The embodiments of this specification can improve the accuracy of container truck anti-lifting recognition to a certain extent.

Description

Technical Field

[0001] The embodiments of the present specification relate to the field of computer vision technology, and more particularly to a method and related device for detecting anti-lifting of a container truck. Background Art

[0002] In the operation of container truck yard bridge or yard, a camera device is usually installed on the bottom bracket of the container truck tire crane to collect the on-site image of the container truck operation and perform anti-lifting detection of the container truck. Specifically, the target is tracked by the motion trajectory of the target object in the collected on-site image. If the target object moves, the container truck may be lifted. In the prior art, a fixed value can be set. When the motion displacement of the target object exceeds the set fixed value, the container truck may be lifted.

[0003] However, the collected on-site images may include a variety of different situations, and fixed numerical settings cannot adapt to all situations. Therefore, fixed numerical values ​​may not be accurate enough to judge whether the container truck is lifted. Summary of the invention

[0004] In view of this, multiple embodiments of this specification are dedicated to providing a container truck anti-lifting detection method and related devices, which can improve the recognition accuracy of container truck anti-lifting to a certain extent.

[0005] Multiple embodiments in this specification provide an anti-lifting detection method for a container truck, the method comprising: detecting an actual spatial attribute value of a designated target object included in the container truck from a reference image collected for the container truck; adaptively setting a target lifting judgment threshold for the container truck based on a conversion relationship between the actual spatial attribute value and a designated target value for the designated target object; and performing anti-lifting detection on the collected on-site image of the container truck based on the target lifting judgment threshold.

[0006] Preferably, the method includes: obtaining an actual distance value between the container truck and the camera device; adjusting the reference lifting judgment threshold value according to a conversion relationship between the actual distance value and the target distance value to obtain the target lifting judgment threshold value.

[0007] Preferably, the method includes: calculating an inverse proportional value between the actual distance value and the target distance value, the inverse proportional value being used as the conversion relationship; multiplying the inverse proportional value by the benchmark lifting judgment threshold to obtain the target lifting judgment threshold.

[0008] Preferably, the method includes: identifying the actual pixel number value of the designated target object in the reference image; wherein the actual pixel number value includes a width pixel number value, a height pixel number value or a total pixel number value; and adjusting the reference lifting judgment threshold according to the conversion relationship between the actual pixel number value and the target pixel number value to obtain the target lifting judgment threshold.

[0009] Preferably, the method includes: collecting historical benchmark images of container trucks; performing multi-category target recognition on the historical benchmark images, and counting the cumulative number of times corresponding to multiple target objects are recognized; and taking the target object whose cumulative number is greater than a specified number threshold as the designated target object.

[0010] Preferably, the method comprises: the designated target object comprises at least one of the following: an object or component existing in all containers, an object or component existing in all container trucks.

[0011] Preferably, the method includes: monitoring the on-site image of the container truck to identify the position change of the designated target object relative to the designated target object in the reference image; when the position change of the designated target object is greater than the target lifting judgment threshold, determining that the container truck is lifted.

[0012] One embodiment of the present specification provides an anti-lifting detection device for a container truck, the anti-lifting detection device comprising: an identification module, used to identify and obtain the actual spatial attribute value of a specified target object in the on-site image data stream collected for the container truck; an adaptive threshold setting module, used to adaptively set a target lifting judgment threshold for the container truck based on a conversion relationship between the actual spatial attribute value and the set target value for the specified target object; and a detection module, used to perform anti-lifting detection on the on-site image stream data corresponding to the container truck based on the target lifting judgment threshold.

[0013] One embodiment of the present specification provides a computer instruction storage medium, wherein the computer storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the anti-lifting detection method for a container truck as described in any one of the above items is implemented.

[0014] One embodiment of the present specification provides a container crane, comprising: an anti-lifting detection device for a container truck for executing any of the methods described above, and a container crane body; the anti-lifting detection device for the container truck is installed on the container crane body.

[0015] The multiple implementation methods provided in this specification detect the actual spatial attribute value of the designated target object included in the container truck from the reference image collected for the container truck; to identify the size of the designated target object in the actual space; based on the conversion relationship between the actual spatial attribute value and the designated target value for the designated target object, adaptively set the target lifting judgment threshold for the container truck; to be applicable to the sensitivity of anti-lifting recognition of the container truck under different environments and conditions; based on the target lifting judgment threshold, anti-lifting detection is performed on the collected on-site image of the container truck, which can improve the recognition accuracy of anti-lifting of the container truck to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic flow chart of an anti-lifting detection method for a container truck is provided for an implementation manner of this specification.

[0017] Figure 2 A schematic diagram of an anti-lifting detection process for a container truck is provided for an implementation manner of this specification.

[0018] Figure 3 A schematic diagram of the position of a container truck anti-lifting detection camera device provided in an implementation manner of this specification.

[0019] Figure 4 A functional module diagram of an anti-lifting detection device for a container truck provided in an embodiment of this specification.

[0020] Figure 5 A schematic diagram of a computer device provided for an embodiment of the present specification. DETAILED DESCRIPTION

[0021] 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 those skilled in the art without creative work are within the scope of protection of this application.

[0022] In the description of the embodiments of the present application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0023] See also Figure 1 and Figure 2 One embodiment of the present specification provides an anti-lifting detection method for a container truck. The anti-lifting detection method can be run in a computer device. The computer device can perform certain data communications with a working condition system at a construction site. The electronic device can receive images captured by a camera device. The anti-lifting detection method can include the following steps.

[0024] Step S110: Detecting actual spatial attribute values ​​of a designated target object included in the container truck from the reference image collected for the container truck.

[0025] In this embodiment, the reference image may include the first frame of an image sequence captured by a camera during container lifting operations. The reference image may be used to determine the motion trajectory of a container truck or container with subsequent image frames. Specifically, for example, an image captured when a spreader starts lifting a container may be used as the reference image.

[0026] In this embodiment, the designated target object may include a specific object identified in the reference image. Based on the identified designated target object, the target lifting judgment threshold for the container truck can be adaptively set. In some embodiments, the designated target object includes at least one of the following: an object or component present in all containers, an object or component present on all container trucks. Specifically, the designated target object may include the container truck as a whole and objects that are widely present on the container. For example, a light source, a container truck as a whole, and a container. In some embodiments, there are some differences between different container trucks, which may result in some non-widespread objects on the container truck, but they are not designated target objects in this application.

[0027] In this embodiment, the actual space attribute value may include the spatial characteristics of the specified target object. The actual space attribute value may be used to represent the size of the specified target object. Specifically, for example, the actual space attribute value of a container may include: a 20-foot container, about 20 feet long, i.e. 6.058 meters; a width of about 8 feet, i.e. 2.438 meters. A 40-foot container, about 40 feet long, i.e. 12.192 meters; a width of about 8 feet, i.e. 2.438 meters. Of course, the actual space attribute value may also be represented based on the number of pixels in the image. Specifically, taking a tire as an example, the width of the tire may be specified as 20 or 30 pixels.

[0028] See also Figure 3 In some embodiments, the camera device may be mounted on the crossbeam on the working side of the container truck of the yard bridge. Figure 3As shown, four cameras can be installed on the beam on the operating side of the container truck of the yard bridge. Among them, the two cameras used to observe the 20-foot container can be set to 2 meters to 4 meters away from the center of the yard bridge, and the two cameras used to observe the 40-foot and 45-foot containers can be set to 5 meters to 7 meters away from the center of the yard bridge. The four cameras can be set to 1.6 meters to 1.8 meters from the ground. The camera angle can be slightly tilted downward at a certain angle to ensure that the frame and the container below 0.5 meters to 1 meter in height can be photographed, for example, it can be tilted downward by 5°-20°.

[0029] Step S120: Adaptively setting a target lifting judgment threshold for the container truck according to a conversion relationship between the actual space attribute value and the designated target value for the designated target object.

[0030] In this embodiment, the specified target value may include a target space attribute value of the specified target object. The specified target value may be used as a reference value for the specified target object. In this way, the target lifting judgment threshold may be adaptively set according to the conversion relationship between the actual space attribute value and the specified target value. Specifically, the specified target value may be the width, height or area along the horizontal direction, and correspondingly, the actual space attribute value may also be the width, height or area. Of course, in some embodiments, the specified target value may also be represented by the number of pixels, such as the number of pixels occupied in the horizontal direction and the number of pixels occupied in the vertical direction, or the total number of pixels in the area involving the specified target object in the image.

[0031] In this embodiment, the designated target value of the target object can be pre-specified. Specifically, the installation position of the camera device can be fixed, and the relative position between the position of the camera device and the position of the captured lane is also fixed. The container truck will be parked in the lane captured by the camera device. Overall, the distance between the container truck and the camera device will be within a numerical range. The container truck can be located in the middle position of the collected image as a reference position, and the designated target value of the part of the container truck that is the designated target object is extracted, for example, the width of the tire is 36 pixels, or the front area includes a total of 460 pixels, etc. The designated target value of the designated target object can be subsequently calculated according to the actual spatial attribute value of the designated target object in the collected scene image, and the conversion relationship between the designated target value and the designated target value. For example, the designated target object is a tire, the width of the tire in the scene image is 25 pixels, and the corresponding designated target value can be 35 pixels. At this time, the conversion relationship can be calculated based on 25 pixels and 35 pixels.

[0032] In this embodiment, the conversion relationship can be a relationship value between the actual space attribute value and the designated target value calculated according to the designated function. Specifically, there may be certain differences in the specific parking positions of different container trucks, so that the distance between different container trucks and the camera device will also change, and then the actual space attribute value of the designated target object in the scene image taken by the camera device will also change accordingly. For example, the farther the distance between the container truck and the camera device, the fewer the number of pixels occupied, and the closer the distance, the more the number of pixels occupied. It can be understood that in the case where the distance between the container truck and the camera device is far, the number of pixels occupied by the designated target object of the container truck in the scene image is also small, so that when the tire in the scene image has a small amount of vertical displacement relative to the position in the reference image, it may be determined that the container truck has been lifted. For example, in the case where the distance between the container truck and the camera device is close, the number of pixels occupied by the designated target object of the container truck in the scene image is also large, so that when the tire in the scene image has a large change in position relative to the reference image, it will be determined that the container truck has been lifted. It can be seen that the distance between the container truck and the camera and the displacement of the designated target object in the scene image relative to the reference image can be inversely related. For example, it is an inverse proportional relationship. Therefore, the inverse proportional value between the actual distance value and the target distance value is calculated, and the inverse proportional value is used as the conversion relationship; the inverse proportional value is multiplied by the reference lifting judgment threshold to obtain the target lifting judgment threshold.

[0033] Specifically, for example, a ratio operation can be performed between the actual spatial attribute value and the specified target value to obtain the ratio value of the two. For example, the specified target object is a tire, the width of the tire in the scene image is 25 pixels, and the corresponding specified target value can be 35 pixels. A ratio operation is performed between 25 and 35 to obtain a ratio value of 5 / 7, and the inverse ratio value 7 / 5 can be used as the conversion relationship.

[0034] In this embodiment, the target lifting judgment threshold for the container truck is adaptively set, which can be understood as dynamically setting the target lifting judgment threshold corresponding to the corresponding container truck according to the collected on-site image. This makes the target lifting judgment threshold more targeted, and thus makes the anti-lifting detection more timely and accurate. In this way, the target lifting judgment threshold can be adaptively set by presetting the benchmark lifting judgment threshold and adjusting the benchmark lifting judgment threshold based on the conversion relationship to obtain the target lifting judgment threshold. Specifically, for example, the benchmark lifting judgment threshold is 18 pixels, and the conversion relationship is 5 / 6. At this time, 18 and 5 / 6 can be multiplied to obtain a target lifting judgment threshold of 15 pixels.

[0035] In some embodiments, the actual space attribute value may include an actual distance value. The anti-lifting detection method for a container truck may further include: obtaining an actual distance value between the container truck and the camera device; and adjusting the reference lifting judgment threshold according to a conversion relationship between the actual distance value and the target distance value to obtain the target lifting judgment threshold.

[0036] In some cases, when a container truck enters the yard crane for operation, it does not always stop at the same distance from the camera device. At different distances, there are certain differences between the images collected by the camera device. In such a case, using only a fixed reference lifting judgment threshold may not be accurate enough.

[0037] In this embodiment, the actual distance value may include the straight-line distance between the container truck and the camera device. The actual distance value between the container truck and the camera device may be measured by a distance measuring device. In some embodiments, the actual distance value between the container truck and the camera device may be calculated by the collected on-site image. Specifically, for example, when the container truck enters the field bridge operation, the camera device obtains the on-site image after the container truck is parked, and the distance value between the container truck and the camera device may be calculated.

[0038] In this embodiment, the target distance value may include a specified reference distance between the camera device and the specified target object of the container truck. For example, when the container truck is parked at a specified position, the distance between the container truck and the camera device is set as the target distance value. Specifically, for example, the tires and frame of the container truck are used as the specified target objects. When the container truck is parked at the standard operating position of the yard bridge, the camera device obtains the distance value between the tires or frame of the container truck as the target distance value.

[0039] In this embodiment, after the actual distance value is measured, the conversion relationship between the actual distance value and the target distance value can be calculated. Specifically, the inverse proportional value of the actual distance value and the target distance value can be used as the conversion relationship. For example, the target distance value is 5 meters and the actual distance value is 7 meters. At this time, the conversion relationship can be obtained as 5 / 7. If the reference lifting judgment threshold is 10 pixels, at this time, according to the conversion relationship, the target lifting judgment threshold is about 7 pixels.

[0040] In some embodiments, the designated target value includes a target pixel number value corresponding to the designated target, and the target pixel number value corresponds to the reference lifting judgment threshold. The anti-lifting detection method for container trucks may also include: identifying the actual pixel number value of the designated target object in the reference image; wherein the actual pixel number value includes a width pixel number value, a height pixel number value, or a total pixel number value; and adjusting the reference lifting judgment threshold according to the conversion relationship between the actual pixel number value and the target pixel number value to obtain the target lifting judgment threshold.

[0041] In this embodiment, the actual pixel number value of the designated target object can be identified in the reference image according to the image recognition algorithm. Specifically, the width pixel number value along the horizontal direction, or the height pixel number value along the vertical direction, or the total pixel number value constituting the designated target object can be identified.

[0042] In this embodiment, the target pixel number value may be the number of pixels occupied by the designated target object in the scene image captured by the camera device when the container truck is located at the designated working position as the target pixel number. Specifically, the target pixel number may also include a target width pixel number value along the horizontal direction, a target height pixel number value along the vertical direction, or a target total pixel number value constituting the designated target object.

[0043] In this embodiment, after the actual pixel number value is obtained, the conversion relationship between the actual pixel number value and the target pixel number value can be calculated according to a preset function. Specifically, for example, the inverse proportional value between the actual pixel number value and the target pixel number value is used as the conversion relationship. The target lifting judgment threshold is obtained by multiplying the reference lifting judgment threshold value with the conversion relationship.

[0044] Step S130: Based on the target lifting judgment threshold, anti-lifting detection is performed on the collected on-site image of the container truck.

[0045] In this embodiment, based on the target lifting judgment threshold, the anti-lifting detection is performed on the collected scene image of the container truck. It can be understood that the scene image can be used to represent the initial state of the container truck. If the distance change between the position of the designated target object in the scene image of the container truck and the position of the designated target object in the reference image exceeds the target lifting judgment threshold, it is determined that the container truck is lifted. Of course, if the distance change between the position of the designated target object in the scene image of the container truck and the position of the designated target object in the reference image does not exceed the target lifting judgment threshold, it is determined that the container truck is not lifted.

[0046] In some embodiments, anti-lifting identification of multiple lanes can also be accomplished by changing the focal length of the camera. Specifically, for example, when the focal length of the camera is small, the collected on-site image of the container truck will be relatively wide and may cover multiple lanes at the same time. In this case, computer vision technology can be used to distinguish container trucks in different lanes and determine whether the container truck is in a lifted state. When the focal length of the camera is large, the collected on-site image of the container truck may be more limited to a single lane, so that the container truck can be identified more accurately and anti-lifting detection can be performed. In combination with the aforementioned embodiments, the target lifting judgment threshold can be dynamically adjusted according to the actual scene requirements.

[0047] In some embodiments, the anti-lifting detection method for container trucks may further include: collecting historical reference images of container trucks; performing multi-category target recognition on the historical reference images, and counting the cumulative number of times corresponding to multiple target objects are recognized; and taking the target object whose cumulative number is greater than a specified number threshold as the designated target object.

[0048] In some cases, by performing multi-category target recognition on historical benchmark images, a cumulative threshold is set, and target objects with cumulative times greater than the specified threshold are determined as designated target objects. This helps to screen out target objects that appear frequently in historical images for subsequent targeted analysis.

[0049] In this embodiment, the historical reference image may include a historical image of the motion state of the container truck captured by the camera device. The historical reference image may provide a reference for the on-site image to analyze the motion state of the multiple target objects.

[0050] In this embodiment, the multi-category target recognition may include recognizing the target object appearing in the historical reference image. The multi-category target may be used to analyze the motion trajectory of the target object. Specifically, for example, the multi-category target recognition may include recognizing container trucks, containers, tires, frames, light sources, etc. in container truck loading and unloading operations.

[0051] In this embodiment, the cumulative number of times the multiple target objects are identified is counted respectively. It can be understood that the target object with the largest cumulative number of appearances in the collected scene image can be used as the designated target object.

[0052] In some embodiments, the anti-lifting detection method for the container truck may further include: monitoring the on-site image of the container truck to identify a position change of a designated target object relative to the designated target object in the reference image; when the position change of the designated target object is greater than the target lifting judgment threshold, determining that the container truck is lifted.

[0053] In some cases, by identifying a designated target object in the scene image and comparing the position with that in the reference image, the movement trajectory of the designated target object can be analyzed and the state change of the container truck can be monitored.

[0054] In this embodiment, the position change of the designated target object relative to the designated target object in the reference image is identified in the monitoring scene image of the container truck. It can be understood that the position of the target object in multiple scene images can be tracked to monitor the movement trajectory of the designated target object relative to the designated target object in the reference image. When the position change of the designated target object is greater than the target lifting judgment threshold, the container truck is deemed to be lifted. Specifically, for example, the frame is taken as the designated target object, and the target lifting judgment threshold is set to 20cm. In the reference image, the frame is 1m away from the ground, which is the initial state. After tracking the trajectory, the frame becomes 2m away from the ground, exceeding the target lifting judgment threshold of 20cm. At this time, it can be considered that the container truck is lifted.

[0055] See also Figure 4 . One embodiment of the present specification also provides an anti-lifting detection device for a container truck. The anti-lifting detection device for a container truck may include an identification module for identifying and deriving the actual spatial attribute value of a designated target object in the on-site image data stream collected for the container truck; an adaptive threshold setting module for adaptively setting a target lifting judgment threshold for the container truck based on a conversion relationship between the actual spatial attribute value and the set target value for the designated target object; and a detection module for performing anti-lifting detection on the on-site image stream data corresponding to the container truck based on the target lifting judgment threshold.

[0056] The functions and effects achieved by the anti-lifting detection device for the container truck can be explained with reference to the aforementioned embodiment and will not be described in detail.

[0057] The embodiments of the present specification also provide a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a computer, the computer executes the anti-lifting detection method for a container truck in any of the above embodiments.

[0058] The embodiments of this specification also provide a computer program product including instructions, which, when executed by a computer, enables the computer to execute the anti-lifting detection method for a container truck in any of the above embodiments.

[0059] See also Figure 5The embodiments of the present disclosure may provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the anti-lifting detection method for a container truck in any of the above embodiments when executing the computer program.

[0060] One embodiment of the present specification provides a container crane, comprising: an anti-lifting detection device for a container truck for executing any of the methods described above, and a container crane body; the anti-lifting detection device for the container truck is installed on the container crane body.

[0061] In some embodiments, the electronic device may include a processor, a non-volatile storage medium, an internal memory, a communication interface, a display device, and an input device connected by a system bus. The non-volatile storage medium may store an operating system and related computer programs.

[0062] It should be understood that the specific examples herein are only intended to help those skilled in the art to better understand the embodiments of the present specification, rather than to limit the scope of the present invention.

[0063] It can be understood that in the various implementations of this specification, the size of the sequence number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the implementation methods of this specification.

[0064] It can be understood that the various embodiments described in this specification can be implemented individually or in combination, and the embodiments of this specification are not limited to this.

[0065] The above is only a specific implementation of this specification, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in this specification, which should be included in the protection scope of this specification. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for detecting anti-lifting of a container truck, characterized in that: The method comprises: Detecting the actual spatial attribute value of the designated target object included in the container truck from the reference image collected for the container truck; wherein the actual spatial attribute value includes the actual distance value between the container truck and the camera device, or the actual pixel number value of the designated target object in the reference image; According to the conversion relationship between the actual space attribute value and the designated target value for the designated target object, the target lifting judgment threshold for the container truck is adaptively set; when the actual space attribute value is the actual distance value, the designated target value is the target distance value; when the actual space attribute value is the actual pixel number value, the designated target value is the target pixel number value; Based on the target lifting judgment threshold, anti-lifting detection is performed on the collected on-site image of the container truck; The step of adaptively setting the target lifting judgment threshold for the container truck according to the conversion relationship between the actual space attribute value and the designated target value for the designated target object includes: According to the conversion relationship between the actual space attribute value and the specified target value, the reference lifting judgment threshold is adjusted to obtain the target lifting judgment threshold; the specified target value corresponds to the reference lifting judgment threshold.

2. The method according to claim 1, characterized in that The actual space attribute value includes the actual distance value; the step of adjusting the reference lifting judgment threshold according to the conversion relationship between the actual space attribute value and the specified target value to obtain the target lifting judgment threshold includes: Calculating an inverse proportional value between the actual distance value and the target distance value, the inverse proportional value being used as the conversion relationship; The target lifting judgment threshold is obtained by multiplying the inverse proportional value by the reference lifting judgment threshold.

3. The method according to claim 1, characterized in that The actual pixel number value includes a width pixel number value, a height pixel number value or a total pixel number value.

4. The method according to claim 1, characterized in that: The method further comprises: Collect historical baseline images of container trucks; Performing multi-category target recognition on the historical reference image, and counting the cumulative number of times each of the multiple target objects has been recognized; The target object whose cumulative number of times is greater than the specified number threshold is used as the specified target object.

5. The method according to claim 4, characterized in that The designated target object includes at least one of the following: an object or component existing in all containers, and an object or component existing in all container trucks.

6. The method according to claim 1, characterized in that Based on the target lifting judgment threshold, the step of performing anti-lifting detection on the collected on-site image of the container truck includes: Monitoring the on-site image of the container truck to identify a position change of a designated target object relative to the designated target object in the reference image; When the position change of the designated target object is greater than the target lifting judgment threshold, it is determined that the container truck is lifted.

7. An anti-lifting detection device for a container truck, characterized in that: The anti-lifting detection device comprises: A recognition module, used to detect the actual spatial attribute value of the designated target object included in the container truck from the reference image collected for the container truck; wherein the actual spatial attribute value includes the actual distance value between the container truck and the camera device, or the actual pixel number value of the designated target object in the reference image; An adaptive threshold setting module, used to adaptively set a target lifting judgment threshold for the container truck according to a conversion relationship between the actual space attribute value and the designated target value for the designated target object; when the actual space attribute value is the actual distance value, the designated target value is the target distance value; when the actual space attribute value is the actual pixel number value, the designated target value is the target pixel number value; A detection module, configured to perform anti-lifting detection on the collected on-site image of the container truck based on the target lifting judgment threshold; Wherein, the adaptive threshold setting module is specifically used for: According to the conversion relationship between the actual space attribute value and the specified target value, the reference lifting judgment threshold is adjusted to obtain the target lifting judgment threshold; the specified target value corresponds to the reference lifting judgment threshold.

8. A computer instruction storage medium, characterized in that: The computer instruction storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.

9. A container crane, characterized in that: The container crane comprises: an anti-lifting detection device for a container truck for executing the method as claimed in any one of claims 1 to 6, and a container crane body; the anti-lifting detection device for a container truck is installed on the container crane body.

Images