Cargo bed stacking detection method and program product

Abstract

The invention discloses a cargo compartment stacking detection method and a program product. The method comprises the steps of collecting left eye compartment inner images and right eye compartment inner images of a target cargo compartment at multiple moments through a binocular camera, and determining an in-compartment depth map of the target cargo compartment at each moment according to the left eye compartment inner images and the right eye compartment inner images at the moments; for each in-compartment depth map, equally dividing the in-compartment depth map into a plurality of article stacking areas, and determining an area average depth corresponding to each article stacking area according to depth values corresponding to a plurality of pixel points in each article stacking area; and a stacking detection result of the target cargo compartment is determined according to the area average depths of the multiple article stacking areas at the multiple moments, wherein the stacking detection result is used for indicating whether the stacking state of the articles stacked in the target cargo compartment is an illegal stacking state or not. According to the technical scheme, the cargo compartment can be detected while being installed, the stacking state of the articles can be detected in time, and the intelligent level of logistics management is effectively improved.

Description

Technical Field

[0001] This invention relates to the field of computer vision technology, and in particular to a method and program product for detecting cargo box stacking. Background Technology

[0002] In the logistics industry, cargo loading and stacking mainly refers to the process of stacking and arranging goods.

[0003] In related technologies, the mainstream inspection of cargo container loading and stacking specifications is usually done manually, mainly relying on visual inspection by management personnel or the use of simple measuring tools. However, manual inspection is inefficient and costly, and cannot achieve efficient and automated inspection of loading and stacking specifications. Summary of the Invention

[0004] This invention provides a method and program for detecting cargo container stacking, in order to solve the technical problems of difficulty in automating cargo container loading and stacking detection and low detection efficiency in related technologies.

[0005] According to one aspect of the present invention, a method for detecting the stacking of cargo compartments is provided, the method comprising:

[0006] The left and right images of the target cargo compartment are captured by a binocular camera at multiple times. For each time, the interior depth map of the target cargo compartment at that time is determined based on the left and right images of the target cargo compartment at that time.

[0007] For each of the compartment depth maps, the compartment depth map is divided into multiple item stacking areas, and the average depth of each item stacking area is determined according to the depth values ​​of multiple pixels in each item stacking area.

[0008] The stacking detection result of the target cargo compartment is determined based on the average depth of the stacking areas of multiple items at multiple times. The stacking detection result is used to indicate whether the stacking state of the items stacked in the target cargo compartment is an illegal stacking state.

[0009] According to one aspect of the present invention, a cargo box stacking detection device is provided, the method comprising:

[0010] The time-deep image acquisition module is used to acquire left and right eye images of the target cargo compartment at multiple times using a binocular camera, and for each time time, determine the interior depth map of the target cargo compartment at that time based on the left and right eye images at that time.

[0011] The stacking area depth determination module is used to divide each of the compartment depth maps into multiple item stacking areas, and determine the average depth of each item stacking area based on the depth values ​​corresponding to multiple pixels in each item stacking area.

[0012] The stacking detection result determination module is used to determine the stacking detection result of the target cargo compartment based on the average depth of the stacking areas of multiple items at multiple times. The stacking detection result is used to indicate whether the stacking state of the items stacked in the target cargo compartment is an illegal stacking state.

[0013] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0014] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform a cargo box stacking detection method according to any embodiment of the present invention.

[0015] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement a cargo box stacking detection method according to any embodiment of the present invention.

[0016] According to another aspect of the present invention, embodiments of this disclosure also provide a computer program product, including a computer program that, when executed by a processor, implements a cargo box stacking detection method as described in any of the embodiments of this disclosure.

[0017] The technical solution of this invention firstly involves acquiring left and right images of the interior of a target cargo compartment at multiple moments using a binocular camera. For each moment, a depth map of the cargo compartment's interior is determined based on the left and right images. This yields depth data at different locations within the cargo compartment at multiple moments, providing a clear understanding of the item stacking situation and facilitating subsequent stacking standard analysis. Secondly, for each depth map, it can be divided into multiple item stacking areas. The average depth of each item stacking area is determined based on the depth values ​​of multiple pixels within each area. By determining the depth data of multiple relatively small areas, refined depth data for each area is obtained, improving the accuracy of item stacking detection. Finally, the stacking detection result of the target cargo compartment is determined based on the average depth of the multiple item stacking areas at multiple moments. This result indicates whether the stacking of items within the target cargo compartment is in an illegal stacking state. By analyzing the stacking detection results, managers can promptly monitor the stacking operations of target vehicles, thereby reducing irregularities during the stacking process and improving the overall stacking utilization rate of the cargo compartment. Therefore, this technical solution, by employing the aforementioned techniques, can quickly, efficiently, and automatically acquire the changing trends of item stacking within the cargo compartment. Through analysis of depth change data in each area, it can promptly detect violations during the stacking process and effectively improve the utilization rate of loading space within the cargo compartment, thus enhancing the economic efficiency of logistics transportation.

[0018] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of a cargo box stacking detection method provided in Embodiment 1 of the present invention;

[0021] Figure 2 This is a flowchart of a cargo box stacking detection method provided according to Embodiment 2 of the present invention;

[0022] Figure 3This is a schematic diagram showing the stacking area of ​​goods in a cargo box stacking detection method according to Embodiment 3 of the present invention;

[0023] Figure 4 This is a schematic diagram of the structure of a cargo box stacking detection device provided in Embodiment 4 of the present invention;

[0024] Figure 5 This is a schematic diagram of the structure of an electronic device that implements the cargo box stacking detection method provided in Embodiment 5 of the present invention. Detailed Implementation

[0025] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "image inside the left eye compartment," "image inside the right eye compartment," "initial item stacking layer," and "target item stacking layer," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0027] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0028] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0029] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0030] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the software or hardware, such as the electronic device, application, server, or storage medium performing the operations of this disclosed technical solution, based on the prompt message.

[0031] As an optional but non-limiting implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0032] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0033] It is understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0034] Example 1

[0035] Figure 1 This invention provides a flowchart of a cargo box stacking detection method according to Embodiment 1. This embodiment is applicable to scenarios in the express logistics industry where cargo box loading and stacking standards are being checked. The method can be executed by a cargo box stacking detection device, which can be implemented in hardware and / or software, optionally through electronic devices such as mobile terminals, PCs, or servers. Figure 1 As shown, the method may specifically include:

[0036] S110. Acquire left and right images of the target cargo compartment at multiple times using a binocular camera. For each time moment, determine the interior depth map of the target cargo compartment at that time based on the left and right images of the target cargo compartment at that time.

[0037] In this context, a binocular camera refers to a imaging system consisting of two cameras or cameras placed at a certain distance, used to acquire depth information of a target location. The target cargo compartment refers to the cargo loading space to be inspected. For example, the target cargo compartment could be the cargo compartment of a vehicle or a warehouse. The left-eye image inside the compartment refers to the image of the interior of the compartment captured by the left camera in the binocular camera system; the right-eye package image refers to the image of the interior of the compartment captured by the right camera in the binocular camera system. The compartment depth map is a depth image generated from the depth information corresponding to each directly visible mail item within the target cargo compartment, or it can refer to a depth image containing the target area generated after filtering out the depth information corresponding to various obstacles.

[0038] Specifically, a binocular camera continuously captures images of the target cargo compartment, obtaining left and right view images of the compartment at multiple moments. These images are then used to construct stereoscopic image pairs at various times. For each moment, based on the stereoscopic image pairs, a stereo matching algorithm is used to calculate the depth information of the surfaces of each mail package within the target cargo compartment, thus determining the compartment's depth map at that moment. This technical solution, by using a binocular camera to determine the depth image of the target cargo compartment at each moment, enables automated detection of changes in the stacking trends of items inside the compartment, timely detection of any irregular stacking behavior, and improves the intelligence level of item stacking detection.

[0039] In real-world scenarios, stacking goods within a cargo compartment may require personnel or equipment. To accurately assess the stacking status, data containing objects (obstacles) other than the goods can be filtered out. In one embodiment, determining the interior depth map of the target cargo compartment at a given time based on the left and right view images of the cargo compartment includes: determining an initial interior depth map of the target cargo compartment at that time based on the left and right view images of the cargo compartment; performing obstacle detection on the initial depth map to obtain obstacle regions; removing the obstacle regions from the initial depth map to obtain the interior depth map of the target cargo compartment at that time.

[0040] This technical solution determines an initial depth map of the target cargo compartment at a given time based on the left and right view images. Obstacle detection is then performed on the initial depth map to identify obstacle regions. These obstacle regions are then removed from the initial depth map to obtain the final depth map of the target cargo compartment at that time. By filtering areas that may affect stacking detection, the accuracy of detecting the stacking status of items can be effectively improved, enabling timely, efficient, and accurate determination of the stacking status of items in the cargo compartment and ensuring efficient use of the cargo compartment's space.

[0041] The initial depth map refers to the raw, unprocessed depth map directly calculated using an algorithm. It typically contains noise and depth information for all objects in the image. Obstacles specifically include at least one of the following: stacking workers, belt conveyors, moving goods, etc. An obstacle region can be understood as the multiple pixel coordinates of each obstacle in the image.

[0042] Specifically, a binocular camera continuously captures images of the target cargo compartment, obtaining left and right view images of the compartment at multiple moments. These images are then used to construct stereo image pairs for each moment. For each moment, based on the stereo image pairs, a stereo matching algorithm determines the initial depth map of all objects within the target cargo compartment at that moment. Obstacles in the initial depth map are identified and their corresponding pixel coordinates are determined, thus identifying obstacle regions. A preset image matting method is used to filter out the depth data corresponding to these obstacle regions, resulting in a depth map of the cargo compartment's interior composed of the remaining objects at the current moment. This technical solution, by filtering out the influence of obstacles on the depth data of stacked items, can accurately obtain the changing trends of each item's stacking area, accurately detect non-standard behaviors of stacking operators during the stacking process, and improve the loading utilization rate of cargo compartment items.

[0043] S120. For each of the compartment depth maps, the compartment depth map is divided into multiple item stacking areas, and the average depth of each item stacking area is determined according to the depth values ​​corresponding to multiple pixels in each item stacking area.

[0044] The item stacking area refers to multiple small areas obtained by dividing the depth map inside the compartment into a preset number of rows and columns. The average depth of the area refers to the average depth value corresponding to the entire item stacking area.

[0045] Specifically, for each depth map inside the compartment, the map is divided into multiple relatively small stacking areas using a region-division method. The depth value at each pixel location within each stacking area is obtained, and the average depth of the stacking area is determined by calculating the average depth of all pixels within that area. By dividing the stacking area into multiple smaller regions, more refined pixel depth data can be obtained, and the accuracy and reliability of the stacking detection results can be improved by observing the changes in depth values ​​between adjacent areas.

[0046] In one embodiment, after determining the average depth of each item stacking area based on the depth values ​​corresponding to multiple pixels within each item stacking area, the method further includes: determining the loading rate of the target cargo box at the time based on the average depth of the multiple item stacking areas at the time, and generating stacking operation prompt information for the target cargo box based on the loading rate.

[0047] Loading rate refers to the proportion of cargo space occupied by a logistics vehicle. Stacking operation prompts are used to indicate problems with previously performed stacking operations and / or stacking operations that need to be performed in the future.

[0048] Specifically, based on the average depth of multiple stacking areas corresponding to multiple items at multiple times, and the pre-acquired maximum average depth data of the cargo compartment bottom, the loading rate of the target cargo compartment at multiple times is determined by dividing the sum of the average depths of the stacking areas corresponding to multiple items at different times by the maximum average depth data of the cargo compartment bottom. Stacking operation prompts for the target cargo compartment are then generated in a timely manner based on the loading rate at different times. The maximum average depth data of the cargo compartment bottom can be obtained by collecting the bottom profile of the target cargo compartment (usually a profile perpendicular to the ground, or referring to the area division in the image of Example 3) before loading begins. This technology calculates the loading rate at multiple times, thus combining the stacking status of items at multiple times to guide the stacking operators in the stacking method in a timely manner, ensuring the stacking utilization rate of the target cargo compartment.

[0049] S130. Determine the stacking detection result of the target cargo compartment based on the average depth of the multiple stacking areas of the items at multiple times. The stacking detection result is used to indicate whether the stacking state of the items stacked in the target cargo compartment is an illegal stacking state.

[0050] Specifically, illegal stacking can include one or more of the following stacking patterns: wall-style stacking, pyramid-style stacking, arch-style stacking, and honeycomb stacking. For example, wall-style stacking typically refers to a structure with gaps between different layers of items. Pyramid-style or arch-style stacking refers to a structure where the bottom layer has a small stacking area, and the upper layers of items continuously protrude outwards. Honeycomb stacking refers to a structure where there are numerous gaps between items. This technical solution is particularly suitable for detecting wall-style stacking.

[0051] Specifically, by comprehensively analyzing the average depth data of multiple item stacking areas at multiple time points, the variation characteristics of the average depth data of each item stacking area over time are determined, thereby determining the stacking detection results of items during the stacking process within the target cargo compartment. This technical solution, through continuous monitoring of the target cargo compartment, effectively combines cargo compartment depth map change data from multiple time points to achieve on-site loading and inspection of items. This ensures the effectiveness of stacking behavior detection, avoids rework due to non-standard stacking, and improves the level of intelligence in logistics item stacking detection.

[0052] In another embodiment, determining the stacking detection result of the target cargo compartment based on the average depth of the multiple item stacking areas at the target time includes: determining multiple initial item stacking layers composed of multiple item stacking areas, wherein the multiple initial item stacking layers are arranged along the height direction of the target cargo compartment; for each initial item stacking layer, determining the maximum average depth and the minimum average depth from the multiple average depths of the multiple item stacking areas in the initial item stacking layer at the target time, determining the target item stacking layer corresponding to the initial item stacking layer based on the difference between the maximum average depth and the minimum average depth; determining the layer average depth of the target item stacking layer based on the multiple average depths of the multiple item stacking areas in the target item stacking layer at the target time, and determining the stacking detection result of the target cargo compartment based on the difference between the layer average depths of the multiple target item stacking layers.

[0053] The initial item stacking layer is a row of stacked items composed of multiple stacked areas. The maximum average depth refers to the average depth of the stacked item area with the highest average depth within a given initial item stacking layer; the minimum average depth refers to the average depth of the stacked item area with the lowest average depth within the same initial item stacking layer. The target item stacking layer is the stacked item layer formed by combining multiple stacked item areas contained within the initial item stacking layer. The layer average depth is the average depth calculated from the average depth data of the multiple stacked item areas within the target item stacking layer.

[0054] Specifically, multiple initial item stacking layers are formed by multiple item stacking areas, arranged along the height of the target cargo compartment. For each initial item stacking layer, the maximum and minimum average depths of the multiple item stacking areas corresponding to these areas are determined at the target time. The target item stacking layer corresponding to the initial item stacking layer is determined based on the difference between the maximum and minimum average depths of the multiple item stacking areas. The average depth of the target item stacking layer is determined based on the average depths of the multiple item stacking areas within the target item stacking layer at the target time. Therefore, the stacking detection result of the target cargo compartment is determined based on the difference in the average depths of the multiple target item stacking layers. This technical solution, by obtaining the target item stacking layers, allows for more flexible calculation of the average depth data for each layer, effectively reducing the amount of data required for computation.

[0055] In one embodiment, determining the target item stacking layer corresponding to the initial item stacking layer based on the difference between the maximum average depth and the minimum average depth includes: determining the initial item stacking layer as the target item stacking layer when the difference between the maximum average depth and the minimum average depth is less than or equal to a preset difference threshold; and dividing the initial item stacking layer into at least two target item stacking layers based on the average depth of the multiple item stacking areas in the initial item stacking layer when the difference between the maximum average depth and the minimum average depth is greater than the preset difference threshold.

[0056] The specific value of the preset difference threshold can be set according to actual needs and is not specifically limited here. For example, the preset difference threshold can be determined based on the size or type of goods stacked in the target cargo compartment.

[0057] Specifically, if the difference between the maximum and minimum average depths in the initial item stacking layer is less than or equal to a preset difference threshold, the initial item stacking layer can be directly identified as the target item stacking layer. However, if the difference between the maximum and minimum average depths in the initial item stacking layer is greater than the preset difference threshold, the average depths of multiple item stacking areas in the initial item stacking layer are clustered to obtain at least two clusters, thereby dividing the initial item stacking layer into at least two target item stacking layers. This technical solution, by classifying the difference between the maximum and minimum average depths, can be applied to more detection scenarios, improving the detection efficiency of item stacking states.

[0058] In one embodiment, after determining the stacking detection result of the target cargo compartment based on the average depth of the multiple stacking areas of the items at multiple times, the method further includes: in response to the stacking detection result indicating that the stacking state of the items stacked in the target cargo compartment is an illegal stacking state, generating violation prompt information corresponding to the illegal stacking state, and displaying the violation prompt information.

[0059] The violation notification information may include at least one of the following: violation area, audible and visual alarm, violation screenshot information, etc., to instruct managers to promptly understand the specific situation of the violation stacking area.

[0060] Specifically, in response to the stacking detection results indicating that the stacked items in the target cargo compartment are in an illegal stacking state, various violation prompts corresponding to the illegal stacking state can be generated immediately and sent to the management personnel terminal in the background. The violation prompts are displayed so that the management personnel can understand the illegal stacking state of the stacking area in a timely manner and continuously monitor the stacking area to prevent operators from continuing to stack items in a non-standard manner.

[0061] In one embodiment, generating violation alert information corresponding to the violation stacking state may include: determining the violation target time when the stacking state of the items stacked in the target cargo compartment is the violation stacking state; determining the violation area among the multiple item stacking areas based on the average depth of the area at the violation target time; and generating violation alert information corresponding to the violation stacking state based on the violation area.

[0062] The "violation target time" refers to the moment when the stacking state of the items is in violation. The "violation area" specifically refers to the image region corresponding to the items that are in a violation stacking state.

[0063] Specifically, the method determines the target time when the stacked items in the target cargo compartment are in an illegal stacking state. Based on the average depth of multiple item stacking areas at the target time, it analyzes the average depth of these areas to identify the illegal areas within the stacked items at that time. Based on these illegal areas, it generates violation alerts corresponding to the illegal stacking state. This technical solution, by comprehensively analyzing the average depth data of multiple item stacking areas at the target time, can more accurately locate the area where illegal stacking occurs, improving the accuracy of illegal stacking area determination and effectively enhancing the efficiency and accuracy of cargo compartment stacking detection.

[0064] The technical solution of this invention firstly involves acquiring left and right images of the interior of a target cargo compartment at multiple moments using a binocular camera. For each moment, a depth map of the cargo compartment's interior is determined based on the left and right images. This yields depth data at different locations within the cargo compartment at multiple moments, providing a clear understanding of the item stacking situation and facilitating subsequent stacking standard analysis. Secondly, for each depth map, it can be divided into multiple item stacking areas. The average depth of each item stacking area is determined based on the depth values ​​of multiple pixels within each area. By determining the depth data of multiple relatively small areas, refined depth data for each area is obtained, improving the accuracy of item stacking detection. Finally, the stacking detection result of the target cargo compartment is determined based on the average depth of the multiple item stacking areas at multiple moments. This result indicates whether the stacking of items within the target cargo compartment is in an illegal stacking state. By analyzing the stacking detection results, managers can promptly monitor the stacking operations of target vehicles, thereby reducing irregularities during the stacking process and improving the overall stacking utilization rate of the cargo compartment. Therefore, this technical solution, by employing the aforementioned techniques, can quickly, efficiently, and automatically acquire the changing trends of item stacking within the cargo compartment. Through analysis of depth change data in each area, it can promptly detect violations during the stacking process and effectively improve the utilization rate of loading space within the cargo compartment, thus enhancing the economic efficiency of logistics transportation.

[0065] Example 2

[0066] Figure 2 This is a flowchart of a cargo compartment stacking detection method provided in Embodiment 2 of the present invention. The solution in this embodiment is a refinement of the technical solution for determining the interior depth map of the target cargo compartment at a given time based on the left and right interior images at the given time, building upon the solutions described in the previous embodiments. Detailed implementation can be found in the description of this embodiment. Technical features that are the same as or similar to those in the previous embodiments will not be repeated here. Figure 2 As shown, the method may specifically include:

[0067] S210. Acquire left and right images of the target cargo compartment at multiple times using a binocular camera. For each time moment, determine the interior depth map of the target cargo compartment at that time based on the left and right images of the cargo compartment at that time.

[0068] S220. For each of the compartment depth maps, the compartment depth map is divided into multiple item stacking areas, and the average depth of each item stacking area is determined according to the depth values ​​corresponding to multiple pixels in each item stacking area.

[0069] S230. For a single item stacking area, determine a depth change curve based on the average depth of the item stacking area in multiple depth maps of the compartment and the time corresponding to the multiple depth maps of the compartment.

[0070] The depth variation curve refers to a curve constructed with time as the horizontal axis and the depth data calculated from the depth information as the vertical axis, showing how depth data changes over time. The target moment can be understood as the moment in the depth variation curve corresponding to a sudden change in depth data.

[0071] Specifically, for a single item stacking area, based on the average depth of the item stacking area in multiple compartment depth maps and the generation time of the multiple compartment depth maps, the depth change curve of the depth data of each item stacking area over time is determined.

[0072] S240. Determine a target time among a plurality of times based on the depth change curves corresponding to the plurality of stacked item areas, and determine the stacking detection result of the target cargo compartment based on the average depth of the plurality of stacked item areas at the target time. The stacking detection result is used to indicate whether the stacking state of the items stacked in the target cargo compartment is an illegal stacking state.

[0073] In this embodiment of the invention, the target time when the depth data undergoes abrupt changes at multiple moments can be determined based on the changes in depth data in the depth variation curves corresponding to multiple stacked item areas. A comprehensive analysis of the average depth of the multiple stacked item areas at the target time is then performed to determine the stacking detection result of the target cargo compartment.

[0074] In one embodiment, determining the target time among the multiple times based on the depth change curves corresponding to the multiple stacked item areas includes: determining the depth abrupt change time of the stacked item area among the multiple times based on the depth change curves corresponding to the stacked item areas; and determining the target time based on the depth abrupt change times corresponding to the multiple stacked item areas.

[0075] Among them, the depth mutation moment refers to the moment when the depth data in the depth change curve shows a significant increase or decrease within adjacent preset moments.

[0076] Specifically, by analyzing the depth variation curves corresponding to multiple stacked item areas, the depth abrupt change times corresponding to sudden increases or decreases in depth data across multiple time points can be determined. By analyzing multiple depth abrupt change times, a target time meeting preset conditions can be identified from at least one depth abrupt change time corresponding to multiple stacked item areas. The preset conditions indicate a sudden change in the depth of the stacked item area, or a significant increase in the depth change rate. For example, the preset conditions can be understood as the difference between the depth data at a certain time and the corresponding depth data at the previous time exceeding a preset depth threshold; or, the difference between the depth data at a certain time and the corresponding depth data at the next time exceeding a preset depth threshold. Further, the preset depth threshold can be determined based on the type or size of the goods loaded in the target cargo compartment. This technical solution, by analyzing multiple depth variation curves and further analyzing multiple depth abrupt change times, can effectively filter out depth abrupt changes caused by certain unexpected factors, thereby obtaining stacking detection results belonging to the illegal stacking state and improving the accuracy of stacking detection in the target cargo compartment.

[0077] In one embodiment, determining the depth abrupt change time of the stacked items at multiple times based on the depth change curve corresponding to the stacked items area includes: determining the depth change rate at multiple times based on the depth change curve corresponding to the stacked items area, determining the depth abrupt change point based on the depth change rate at multiple times, and determining the time corresponding to the depth abrupt change point as the depth abrupt change time.

[0078] The depth change rate is used to indicate how quickly depth data changes over time.

[0079] Optionally, the depth change curves corresponding to multiple stacked item areas are differentiated to obtain the first derivative of the depth change curves at multiple time points, i.e., the depth change rate. The depth change rate at multiple time points is determined, and the maximum value among the depth change rates is identified as the depth abrupt change point. The time corresponding to the depth abrupt change point is determined as the depth abrupt change time. This technical solution can efficiently and easily determine the depth abrupt change time by calculating the derivative of the depth change curve, thereby improving the calculation efficiency for the target time.

[0080] In one embodiment, determining the target time based on the depth abrupt change times corresponding to the multiple stacked item areas may include: determining all the depth abrupt change times corresponding to the multiple stacked item areas as the target time. By determining all the depth abrupt change times corresponding to the multiple stacked item areas as the target time, the determined target time can be more comprehensive, which can improve detection efficiency and refine the detection granularity as much as possible, thereby improving the accuracy and timeliness of detection.

[0081] In another embodiment, determining the target time based on the depth abrupt change times corresponding to the multiple item stacking areas may include: when the depth abrupt change times are determined by the multiple item stacking areas and the multiple item stacking areas corresponding to the depth abrupt change times are consecutive, determining the depth abrupt change time as the target time. This technical solution, by combining the depth abrupt change times of multiple item stacking areas, can effectively identify the target time that needs to be identified and detected subsequently, avoiding unnecessary detection, reducing detection performance consumption, and improving the detection efficiency of item stacking status. This technical solution is particularly suitable for scenarios where the divided item stacking areas are relatively small.

[0082] This technical solution, through the construction of a depth change curve and based on a determined target time, can effectively and accurately locate the internal depth image frame where the stacking state of items is abnormal. By analyzing the average depth data of each item stacking area in the internal depth image at the target time, the efficiency of determining the stacking detection results is improved.

[0083] Example 3

[0084] Embodiment 3 of the present invention provides a schematic diagram of the item stacking area in a cargo box stacking detection method. To better illustrate the technical solution provided by this embodiment, this embodiment uses "wall-style stacking" as an example. This embodiment is illustrated by the following steps, and the schematic diagram of the item stacking area in this embodiment is shown below. Figure 3 As shown, specific implementation methods can be found in the description of this embodiment. Technical features that are the same as or similar to those in the foregoing embodiments will not be repeated here.

[0085] 1. Camera installation and deployment:

[0086] A calibrated smart binocular camera is fixedly installed inside the loading pallet to ensure that its field of view covers the entrance of the cargo compartment and the four corners of the bottom side of the cargo compartment. The important parameters of the binocular camera are calibrated and corrected to ensure the quality of image acquisition.

[0087] 2. Image Acquisition and Depth Map Generation

[0088] During the loading process, the binocular camera simultaneously captures two images from the left and right sides, forming a stereoscopic image pair.

[0089] The Semi-Global Block Matching (SGBM) algorithm is used to calculate the depth map of the cargo compartment's interior by analyzing the stereo image pairs. Each pixel value in the depth map represents the distance between that point and the camera.

[0090] 3. Depth profile extraction

[0091] 1) Extract the depth map information inside the cargo compartment, and identify obstacles such as workers, belt conveyors, and moving goods by marking them, and extract the depth value corresponding to the obstacle part.

[0092] 2) The depth map inside the cargo compartment is divided into multiple equal regions, and areas corresponding to obstacles such as personnel and conveyor belts are excluded by image cutout. The depth values ​​of the remaining equal regions are calculated, and the average depth value of all pixels in each equal region is taken as the depth value of a single equal region. Specifically, the display diagram of the item stacking area in this technical solution can be as follows: Figure 3 As shown, the average depth data of the item stacking area is obtained by calculating the depth data corresponding to each pixel in each equally divided item stacking area.

[0093] 4. Analysis of the quality characteristics of cargo box stacking and judgment of "stacking" violations

[0094] The illegal operation of the "code wall" is specifically manifested in the rapid change of the depth of the scanning area corresponding to each stack of items in a short period of time, and in a vertical comparison, a clear step edge is formed on the depth map inside the cargo compartment.

[0095] The first part involves dividing the cargo compartment into equal regions based on the depth map of the cargo compartment at multiple time points. Based on the average depth value and time of the multiple equal regions, as well as the preset depth change curve, the first derivative (i.e., the depth change rate) of the depth change curve is determined. By determining the maximum value of the absolute value of the derivative, the depth mutation point and the mutation time corresponding to the depth mutation point are determined.

[0096] The second part analyzes the difference between the average depths of each row at the same time (the moment of sudden change).

[0097] Given multiple item stacking layers composed of multiple equally divided regions, with each item stacking layer being a row, for a given item stacking layer, obtain the maximum and minimum depth values ​​of each equally divided region within that item stacking layer:

[0098] 1) If the first difference between the maximum depth value and the minimum depth value does not exceed the first preset distance, then calculate the average depth M1 of the stacking layer of the item based on the depth values ​​of each equally divided area in the stacking layer of the item;

[0099] 2) If the first difference between the maximum depth value and the minimum depth value exceeds the first preset distance, then depth clustering is performed on each equally divided area in the stacking layer of the item to divide the multiple equally divided areas in the stacking layer of the item into two clusters, and the average depth A1 and B1 of the multiple equally divided areas in the two clusters are calculated respectively.

[0100] Based on the above method, calculate the depth data of multiple rows (item stacking layers).

[0101] Regarding 1), based on the multiple depth average data corresponding to multiple rows (item stacking layers), determine the second difference between the maximum depth average Nmax and the minimum depth average Nmin in multiple rows. If the second difference is greater than the second preset distance, it can be considered that at the same moment (abrupt moment), there is a "wall" stacking at the target position of the depth map inside the cargo compartment.

[0102] Regarding point 2), based on the average depth values ​​A1, A2, A3, etc., and average depth values ​​B1, B2, B3, etc., corresponding to two clusters in multiple rows (item stacking layers), a first depth sequence can be constructed by selecting the larger average depth value in each row, and a second depth sequence can be constructed using the other relatively smaller average depth value. A third difference between Amax and Amin is calculated by selecting the maximum average depth value Amax and the minimum average depth value Amin in the first depth sequence; and a fourth difference between the maximum average depth value Bmax and the minimum average depth value Bmin is calculated in the second depth sequence. If at least one of the third and fourth differences is greater than a preset third preset distance, or under other preset conditions (such as processing depth data from the same side, processing only the larger depth data in the same row, etc.), it can be considered that at the same moment (the moment of sudden change), the target location of the depth map inside the cargo compartment exhibits a wall-like stacking.

[0103] 5. Results Output and Early Warning

[0104] If illegal stacking (wall-style stacking) is detected, the system will immediately trigger an early warning mechanism. The illegal area can be outlined in the image on the cargo loading and unloading terminal screen, an audible and visual alarm can be issued, and screenshots and messages of the violation can be pushed to the mobile terminal of the management personnel, so that the management personnel can pay attention to the loading status of the cargo box in a timely manner.

[0105] This technical solution, by employing the aforementioned methods, enables "inspection while loading" based on real-time depth data collection, eliminating unauthorized stacking practices at the source and avoiding the waste of time and labor costs caused by rework of express items. The inspection process requires no manual intervention, effectively reducing the workload of staff and improving the level of intelligence in logistics management.

[0106] Example 4

[0107] Figure 4 This is a schematic diagram of a cargo box stacking detection method provided in Embodiment 4 of the present invention. Figure 4 As shown, the device includes: a time-depth image acquisition module 401, a stacking area depth determination module 402, and a stacking detection result determination module 403.

[0108] The module 401 is used to acquire left and right images of the target cargo compartment at multiple times using a binocular camera. For each time moment, it determines the cargo compartment depth map based on the left and right images. The stacking area depth determination module 402 is used to divide each cargo compartment depth map into multiple stacking areas and determine the average depth of each stacking area based on the depth values ​​of multiple pixels within each stacking area. The stacking detection result determination module 403 is used to determine the stacking detection result of the target cargo compartment based on the average depth of the stacking areas at multiple times. The stacking detection result indicates whether the stacking status of the items in the target cargo compartment is an illegal stacking status.

[0109] The technical solution of this invention involves the following steps: First, the time-based depth image acquisition module 401 acquires left and right images of the target cargo compartment at multiple time points using a binocular camera. For each time point, it determines the cargo compartment's depth map based on the left and right images. This provides depth data at different locations within the cargo compartment at multiple time points, allowing for a clear understanding of the item stacking situation and facilitating subsequent stacking standard analysis. Second, the stacking area depth determination module 402 divides each cargo compartment depth map into multiple item stacking areas. It then determines the average depth of each item stacking area based on the depth values ​​of multiple pixels within each area. By determining the depth data of multiple relatively small areas, refined depth data for each area is obtained, improving the accuracy of item stacking detection. Finally, the stacking detection result determination module 403 can determine the stacking detection result of the target cargo compartment based on the average depth of the stacking areas of multiple items at multiple times. The stacking detection result is used to indicate whether the stacking state of the items stacked in the target cargo compartment is an illegal stacking state. Through the stacking detection result, managers can promptly monitor the stacking operations of the target vehicle, thereby reducing various irregular behaviors during the stacking process and improving the overall stacking utilization rate of the cargo compartment. Therefore, this technical solution, by adopting the above-mentioned technical means, can quickly, efficiently, and automatically obtain the changing trend of item stacking in the cargo compartment, and by analyzing the depth change data of each area, promptly detect illegal behaviors during the stacking process, effectively improve the utilization rate of the loading space in the cargo compartment, and thus improve the economic efficiency of logistics transportation.

[0110] Based on the above-mentioned optional technical solutions, the stacking detection result determination module 403 may optionally include: a depth change curve determination unit and a stacking detection result determination unit. The depth change curve determination unit is used to determine a depth change curve for a single item stacking area based on the average depth of the item stacking area in multiple depth maps of the compartment and the corresponding time in the multiple depth maps of the compartment. The stacking detection result determination unit is used to determine a target time among multiple times based on the depth change curves corresponding to multiple item stacking areas, and to determine the stacking detection result of the target compartment based on the average depth of the multiple item stacking areas at the target time.

[0111] Based on the above-mentioned optional technical solutions, the stacking detection result determination unit may optionally include: a depth abrupt change time determination unit and a target time determination unit. The depth abrupt change time determination unit is used to determine the depth abrupt change time of the stacked item area in multiple time intervals based on the depth change curve corresponding to the stacked item area; the target time determination unit is used to determine a target time based on the depth abrupt change times corresponding to the multiple stacked item areas.

[0112] Based on the above-mentioned optional technical solutions, the depth mutation time determination unit may optionally include a depth mutation point determination unit. The depth mutation point determination unit is configured to determine the depth change rate at multiple times based on the depth change curves corresponding to the multiple stacked item areas, determine the depth mutation point based on the depth change rate at the multiple times, and determine the time corresponding to the depth mutation point as the depth mutation time.

[0113] Based on the above-mentioned optional technical solutions, the stacking detection result determination unit may optionally include: an initial item stacking layer determination unit, a target item stacking layer determination unit, and a layer average depth determination unit. The initial item stacking layer determination unit is used to determine multiple initial item stacking layers composed of multiple item stacking areas, wherein the multiple initial item stacking layers are arranged along the height direction of the target cargo compartment; the target item stacking layer determination unit is used to, for each initial item stacking layer, determine the maximum and minimum average depths from multiple average depths of the multiple item stacking areas in the initial item stacking layer at the target time, and determine the target item stacking layer corresponding to the initial item stacking layer based on the difference between the maximum and minimum average depths; the layer average depth determination unit is used to determine the layer average depth of the target item stacking layer based on the multiple average depths of the multiple item stacking areas in the target item stacking layer at the target time, and determine the stacking detection result of the target cargo compartment based on the difference between the layer average depths of the multiple target item stacking layers.

[0114] Based on the above-mentioned optional technical solutions, the target item stacking layer determining unit may optionally include: a first target item stacking layer determining unit and a second target item stacking layer determining unit. The first target item stacking layer determining unit is used to determine the initial item stacking layer as a target item stacking layer when the difference between the maximum average depth and the minimum average depth is less than or equal to a preset difference threshold. The second target item stacking layer determining unit is used to divide the initial item stacking layer into at least two target item stacking layers based on the average depth of the multiple item stacking areas in the initial item stacking layer when the difference between the maximum average depth and the minimum average depth is greater than the preset difference threshold.

[0115] Based on the above-mentioned optional technical solutions, the time-based depth image acquisition module 401 may optionally include an obstacle removal unit. The obstacle removal unit is configured to determine an initial depth map of the target cargo compartment at the given time based on the left and right view images of the cargo compartment at the given time, perform obstacle detection on the initial depth map to obtain obstacle regions, remove the obstacle regions from the initial depth map, and obtain the cargo compartment depth map of the target cargo compartment at the given time.

[0116] Based on the above-mentioned optional technical solutions, the cargo box stacking detection device may optionally include: a first prompt information generation module. The prompt information generation module is used to determine the loading rate of the target cargo box at the given time based on the average depth of the areas corresponding to the multiple stacking areas of the items at the given time, and to generate stacking operation prompt information for the target cargo box based on the loading rate.

[0117] Based on the above-mentioned optional technical solutions, the cargo compartment stacking detection device may optionally include: a second prompt information generation module. The second prompt information generation module is used to generate and display violation prompt information corresponding to the violation stacking state in response to the stacking detection result indicating that the stacking state of the items stacked in the target cargo compartment is an illegal stacking state.

[0118] The cargo container stacking detection device provided in this embodiment of the invention can execute a cargo container stacking detection method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing a cargo container stacking detection method. Technical details not described in detail in this embodiment can be found in any of the cargo container stacking detection methods described in this embodiment of the invention.

[0119] Example 5

[0120] Figure 5 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0121] like Figure 5As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0122] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0123] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as a cargo box stacking detection method.

[0124] In some embodiments, a cargo box stacking detection method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the cargo box stacking detection method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform a cargo box stacking detection method by any other suitable means (e.g., by means of firmware).

[0125] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0126] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0127] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0128] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0129] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0130] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0131] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication unit 19, or installed from storage unit 18, or installed from ROM 12. When the computer program is executed by processor 11, it performs the functions defined in the methods of the embodiments of the present invention.

[0132] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0133] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for detecting the stacking of cargo compartments, characterized in that, include: The left and right images of the target cargo compartment are captured by a binocular camera at multiple times. For each time, the interior depth map of the target cargo compartment at that time is determined based on the left and right images of the target cargo compartment at that time. For each of the compartment depth maps, the compartment depth map is divided into multiple item stacking areas, and the average depth of each item stacking area is determined according to the depth values ​​of multiple pixels in each item stacking area. The stacking detection result of the target cargo compartment is determined based on the average depth of the stacking areas of multiple items at multiple times. The stacking detection result is used to indicate whether the stacking state of the items stacked in the target cargo compartment is an illegal stacking state.

2. The cargo box stacking detection method according to claim 1, characterized in that, The step of determining the stacking detection result of the target cargo compartment based on the average depth of the stacking areas of multiple items at multiple times includes: For a single item stacking area, a depth change curve is determined based on the average depth of the item stacking area in multiple depth maps of the compartment and the time corresponding to the multiple depth maps of the compartment; The target time among the multiple time points is determined based on the depth change curves corresponding to the multiple stacked item areas, and the stacking detection result of the target cargo box is determined based on the average depth of the multiple stacked item areas at the target time.

3. The cargo box stacking detection method according to claim 2, characterized in that, Determining the target time among the multiple time points based on the depth change curves corresponding to the multiple stacked item areas includes: The depth change time of the stacked item area at multiple times is determined based on the depth change curve corresponding to the stacked item area. The target time is determined based on the depth mutation time corresponding to the multiple stacked areas of the items.

4. The cargo box stacking detection method according to claim 3, characterized in that, Determining the depth abrupt change moments of the stacked item area at multiple moments based on the depth change curve corresponding to the stacked item area includes: Based on the depth change curves corresponding to the stacked areas of the items, the depth change rate at multiple times is determined, and the depth mutation point is determined based on the depth change rate at multiple times. The time corresponding to the depth mutation point is determined as the depth mutation time.

5. The cargo box stacking detection method according to claim 2, characterized in that, The step of determining the stacking detection result of the target cargo compartment based on the average depth of the multiple stacking areas of the items at the target time includes: Determine multiple initial item stacking layers, which consist of multiple item stacking areas, and arrange the multiple initial item stacking layers along the height direction of the target cargo box; For each initial item stacking layer, the maximum average depth and the minimum average depth are determined from the average depths of multiple item stacking areas in the initial item stacking layer at the target time, and the target item stacking layer corresponding to the initial item stacking layer is determined based on the difference between the maximum average depth and the minimum average depth. The average depth of the target item stacking layer is determined based on the average depth of multiple stacking areas in the target item stacking layer at the target time, and the stacking detection result of the target cargo box is determined based on the difference between the average depths of the multiple target item stacking layers.

6. The cargo box stacking detection method according to claim 5, characterized in that, Determining the target item stacking layer corresponding to the initial item stacking layer based on the difference between the maximum average depth and the minimum average depth includes: If the difference between the maximum average depth and the minimum average depth is less than or equal to a preset difference threshold, the initial item stacking layer is determined as the target item stacking layer. If the difference between the maximum average depth and the minimum average depth is greater than a preset difference threshold, the initial item stacking layer is divided into at least two target item stacking layers according to the average depth of the multiple item stacking areas in the initial item stacking layer.

7. The cargo box stacking detection method according to claim 1, characterized in that, Determining the interior depth map of the target cargo compartment at the specified time based on the left and right view images at the specified time includes: Based on the left and right images inside the cargo compartment at the specified time, an initial depth map of the cargo compartment inside the target cargo compartment at the specified time is determined. Obstacle detection is performed on the initial depth map to obtain obstacle regions. The obstacle regions in the initial depth map are removed to obtain the depth map inside the target cargo compartment at the specified time.

8. The cargo box stacking detection method according to claim 1, characterized in that, After determining the average depth of each item stacking area based on the depth values ​​corresponding to multiple pixels within each item stacking area, the method further includes: The loading rate of the target cargo box at the time is determined based on the average depth of the areas corresponding to the multiple stacking areas of the items at the time, and stacking operation prompt information of the target cargo box is generated based on the loading rate.

9. The cargo box stacking detection method according to claim 1, characterized in that, After determining the stacking detection result of the target cargo compartment based on the average depth of the multiple stacking areas of the items at multiple times, the method further includes: In response to the stacking detection result indicating that the stacking status of the items stacked in the target cargo compartment is an illegal stacking status, a violation prompt message corresponding to the illegal stacking status is generated and displayed.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the cargo box stacking detection method as described in any one of claims 1-9.

Images