Container yard location determination method, device, equipment and storage medium

Abstract

The present invention discloses a method, device, equipment and storage medium for determining the location of containers in a container yard. The method includes: constructing a three-dimensional model of a target yard; acquiring multiple container images of a target area; stitching each image to obtain a panoramic image of the target area; detecting recognizable containers in the panoramic image and determining whether there are unrecognizable containers in the target area; if not, executing a first solution; if so, executing the first solution for recognizable containers and executing a second solution for unrecognizable containers; the first solution is to determine the location of recognizable containers based on the mapping relationship between the three-dimensional model and the panoramic image; the second solution is to determine the location of unrecognizable containers based on the entry and exit information of containers in the target area, the type of container cargo and the shipping and receiving locations, the recognizable container information and the stacking location information of which the image cannot be collected. The present application can quickly determine the location of each container in the yard with high accuracy.

Description

Technical Field

[0001] The present invention relates to the field of electromagnetic characteristic calculation technology, and in particular to a method, device, equipment and storage medium for determining a container location in a container yard. Background Art

[0002] With the rapid development of logistics, the number of containers in container yards and the frequency of their entry and exit have increased dramatically. To ensure orderly container storage, timely inventory and data updates are essential. Any change in a container's location requires prompt recording. However, manual record keeping is not only inefficient but also prone to errors. This can lead to situations where the correct container location cannot be found or a newly arrived container's location is occupied, severely impacting normal yard operations.

[0003] Therefore, there is an urgent need for a method, device, equipment and storage medium for determining container space in a container yard. Summary of the Invention

[0004] The present invention provides a method, device, equipment, and storage medium for determining the location of containers in a container yard, which can quickly determine the location of each container in the yard with high accuracy. The technical solution is as follows:

[0005] In one aspect, a method for determining a container location in a container yard is provided, the method comprising:

[0006] Constructing a three-dimensional model of a target container yard; the three-dimensional model includes a plurality of container areas, each of the container areas includes a plurality of stacking locations, and each of the stacking locations is used to store containers;

[0007] Acquire multiple container images of a target area, where the target area is at least a portion of the target yard;

[0008] Stitching each of the container images to obtain a panoramic image of the target area;

[0009] Detecting recognizable containers in the panoramic image based on a pre-trained detection model and determining whether there are unrecognizable containers in the target area; if not, executing the first solution; if so, executing the first solution for recognizable containers and executing the second solution for unrecognizable containers;

[0010] The first solution is: determining the location of each identifiable container based on a mapping relationship between the three-dimensional model and the panoramic image;

[0011] The second solution is to determine the location of each unrecognizable container based on the entry and exit information of the containers in the target area, the container cargo type and the shipping and receiving locations, the identifiable container information and the stacking information of the uncaptured images.

[0012] In another aspect, a device for determining a container location in a container yard is provided, the device comprising:

[0013] A construction unit is used to construct a three-dimensional model of the target container yard; the three-dimensional model includes a plurality of container areas, each of the container areas includes a plurality of stacking locations, and each of the stacking locations is used to store containers;

[0014] an acquisition unit, configured to acquire a plurality of container images of a target area, wherein the target area is at least a portion of the target yard;

[0015] a stitching unit, configured to stitch each of the container images to obtain a panoramic image of the target area;

[0016] a judgment unit, configured to detect recognizable containers in the panoramic image based on a pre-trained detection model, and to judge whether there are unrecognizable containers in the target area;

[0017] a first determining unit, configured to execute a first solution for the identifiable containers in the target area, wherein the first solution is to determine a location of each identifiable container in the target area based on a mapping relationship between the three-dimensional model and the panoramic image;

[0018] The second determination unit is used to execute a second solution for the unrecognizable containers in the target area, and the second solution is: based on the entry and exit information of the containers in the target area, the type of container cargo and the shipping and receiving locations, the recognizable container information and the stacking information of the unrecognizable containers in the target area, determine the container location of each unrecognizable container in the target area.

[0019] On the other hand, a computer device is provided, which includes a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to implement the steps of the above-mentioned method for determining a container yard location.

[0020] On the other hand, a computer-readable storage medium is provided, wherein a computer program is stored in the storage medium, and when the computer program is executed by a processor, the steps of the above-mentioned method for determining a container location in a container yard are implemented.

[0021] On the other hand, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the steps of the above-mentioned method for determining a container location in a container yard are implemented.

[0022] An embodiment of the present invention provides a method for determining the location of a container yard. First, a three-dimensional model of the target yard is constructed. Then, when it is necessary to inventory a target area in the target yard, an image of the target area is captured, a panoramic image of the target area is spliced ​​together, and the containers in the panoramic image are identified. For identifiable containers, the location of each container is determined based on the mapping relationship between the three-dimensional model and the panoramic image. For unidentifiable containers, the location of each container is determined based on the entry and exit information of the container in the target area, the container cargo type and the shipping and receiving locations, the identifiable container information, and the stacking information for which images cannot be captured. In this way, the location of each container can be determined through three-dimensional modeling, image recognition, and information such as the entry and exit information and cargo type of the container. This application does not require manual operation and can quickly determine the location of each container in the yard with high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0024] Figure 1 This is a flow chart of a method for determining a container location in a container yard provided by one embodiment of the present invention;

[0025] Figure 2 This is a structural diagram of a container yard location determination device provided by one embodiment of the present invention;

[0026] Figure 3 This is a hardware architecture diagram of a computer device provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0028] The specific implementation of the above concept is described below.

[0029] Please refer to Figure 1 , an embodiment of the present invention provides a method for determining a container location in a container yard, the method comprising:

[0030] Step 100: construct a three-dimensional model of the target container yard; the three-dimensional model includes multiple container areas, each container area includes multiple stacking locations, and each stacking location is used to store containers;

[0031] Step 102: Acquire multiple container images of a target area, where the target area is at least a portion of a target container yard;

[0032] Step 104: stitching each container image to obtain a panoramic image of the target area;

[0033] Step 106 , detecting recognizable containers in the panoramic image based on the pre-trained detection model, and determining whether there are unrecognizable containers in the target area;

[0034] Step 108: executing a first solution for the identifiable containers in the target area. The first solution is: determining the location of each identifiable container based on a mapping relationship between the three-dimensional model and the panoramic image;

[0035] Step 110, executing the second solution for the identifiable containers in the target area, the second solution is: based on the entry and exit information of the containers in the target area, the container cargo type and the delivery and receipt locations, the identifiable container information and the stacking information of the uncaptured images, determine the container location of each unidentifiable container.

[0036] In an embodiment of the present invention, a three-dimensional model of the target yard is first constructed. Then, when it is necessary to inventory the target area in the target yard, a panoramic image of the target area can be spliced ​​together by capturing images of the target area, and the containers in the panoramic image can be identified. For identifiable containers, the location of each container is determined based on the mapping relationship between the three-dimensional model and the panoramic image; for unidentifiable containers, the location of each container is determined based on the entry and exit information of the containers in the target area, the type of container cargo and the shipping and receiving locations, the identifiable container information, and the stacking information for which images cannot be captured. In this way, the location of each container can be determined through three-dimensional modeling, image recognition, and information such as the entry and exit of the container and the type of cargo. This application does not require manual operation and can quickly determine the location of each container in the yard with high accuracy.

[0037] Described below Figure 1 How to perform the steps shown.

[0038] First, for step 100, a three-dimensional model of the target storage yard is constructed, including:

[0039] Collect geographic coordinate information of each container area in the target yard;

[0040] Based on the geographic coordinate information of each container area, a three-dimensional engine is used to perform three-dimensional modeling of the target yard to obtain a three-dimensional model of the target yard, and the correspondence between the three-dimensional model and the actual geographic coordinates is determined based on the three-dimensional map.

[0041] In this step, the geographic coordinates of each container area are determined using a high-precision positioning device, mainly including the length, width, height, and latitude and longitude of each corner point. By determining the correspondence between the 3D model and the actual geographic coordinates, a basis for subsequent container location identification can be provided.

[0042] With respect to step 102 , a plurality of container images of a target area are acquired, where the target area is at least a portion of a target yard.

[0043] In this step, the target area may be one container area or multiple container areas in the target yard, or the entire target yard, and its specific range is determined according to user needs.

[0044] In some embodiments, each container image is captured by an image acquisition device, which is mounted on a mobile platform. The mobile platform moves along a pre-planned path, and a plurality of acquisition points are provided on the pre-planned path. The image acquisition device captures the corresponding container image at each acquisition point according to a preset angle.

[0045] In this step, the mobile platform can be a drone, vehicle, or automated guided vehicle (AGV), adapting to different yard sizes and layouts. By using a mobile platform, the image acquisition device captures images at each point at a set angle, covering the entire target yard area and ensuring that every container is identified. Furthermore, a 30%-50% overlap is ensured between adjacent images, and the location information of the sample corresponding to each image is recorded.

[0046] In addition, the mobile platform is equipped with a fill light device and processor. This fill light device ensures high-quality container images in a variety of environmental conditions (such as varying weather and lighting conditions). The device processor integrates edge computing capabilities on the mobile platform to perform pre-processing operations such as image denoising and color correction. It then uses a deep learning algorithm to identify the container number and sends the identified container number and image coordinate information to the client computer, reducing data transmission and improving system response speed.

[0047] In some embodiments, the pre-planned path is determined based on the target yard's layout, aisle width, obstacle locations, target area traffic level, and container turnover rate. For example, in areas with high container turnover, the mobile platform needs to make frequent trips; conversely, in areas with low container turnover, the mobile platform's travel frequency is reduced. Similarly, the frequency with which the image acquisition device captures images at each acquisition point is determined by the container turnover rate corresponding to that point. If the container turnover rate at that point is high, the number of acquisitions should be increased; otherwise, the number should be reduced.

[0048] With respect to step 104 , each container image is stitched together to obtain a panoramic image of the target area.

[0049] In this step, the scale-invariant feature transform matching algorithm (SIFT) is used to extract feature points in the image, the fast nearest neighbor search package (FLANN) is applied for feature matching, and the multi-band fusion technology is applied using the estimated homography matrix (RANSAC, Random sample consensus) to generate a seamless panoramic image of the target area.

[0050] With respect to step 106 , the recognizable containers in the panoramic image are detected based on the pre-trained detection model, and it is determined whether there are unrecognizable containers in the target area.

[0051] In this step, the pre-trained detection model can be a model such as YOLOv5. Image recognition models are widely used in various fields, and the model structure and training process will not be described here.

[0052] In some embodiments, determining whether there is an unrecognizable container in the target area is based on a first formula:

[0053] The first formula is: Y = A + CDB

[0054] If Y = 0, it is determined that there are no unrecognizable containers in the target area;

[0055] If Y ≥ 1, it is determined that there are unrecognizable containers in the target area;

[0056] Where Y is the number of unrecognizable containers in the target area; A is the number of containers in the target area before the current image acquisition; C and D are the number of containers entering and leaving the target area between the current image acquisition and the previous image acquisition, respectively; B is the number of recognizable containers in the panoramic image corresponding to the current image acquisition; Y, A, B, C, and D are all natural numbers.

[0057] With respect to step 108, based on the mapping relationship between the three-dimensional model and the panoramic image, determining the location of each identifiable container includes:

[0058] Based on the position and angle of the acquisition point where the image acquisition device is located when acquiring the panoramic image, a conversion relationship between the coordinate system of the panoramic image and the geographic coordinate system is established to generate a GeoTIFF file with geographic information;

[0059] Based on the panoramic image and the GeoTIFF file, a preset algorithm is used to perform 3D mesh reconstruction and texture mapping on the panoramic image to obtain the coordinates of each identifiable container in the panoramic image;

[0060] Calculating the actual geographic coordinates of each identifiable container in the panoramic image based on the coordinates of each identifiable container in the panoramic image and the correspondence between the three-dimensional model and the actual geographic coordinates;

[0061] The actual geographic coordinates of each identifiable container in the panoramic image are compared with the container location layout database of the target yard to obtain the location of each identifiable container.

[0062] In this embodiment, the SfM algorithm is used to estimate the position and posture of the image acquisition device, the MVS algorithm is applied to generate a dense point cloud, and the panoramic image is meshed and texture mapped.

[0063] With respect to step 110, based on the container entry and exit information, container cargo type and shipping and receiving locations, identifiable container information, and stack location information for which images cannot be captured in the target area, the location of each unidentifiable container is determined, including:

[0064] Determine a set of stacking locations where images cannot be captured based on a target area's bin layout database and a result of determining the bin locations of the target area before the current image capture.

[0065] Calculating the number of unrecognizable containers in the target area based on the first formula;

[0066] Based on the container number of each container in the target area before the current image acquisition, the container numbers of containers entering and leaving the target area between the current image acquisition and the previous image acquisition, and the container numbers of identifiable containers in the panoramic image corresponding to the current image acquisition, the container number set corresponding to the unidentifiable containers in the target area is determined; the container number of each container in the target area is unique, and the corresponding container's cargo type, entry and exit time, and shipping address are found by the container number;

[0067] Based on the loading and unloading rules of the target area, the cargo type corresponding to each container number, the entry and exit time and the shipping and receiving address, a stacking location is determined for each container number from the stacking location set where images cannot be collected, and the container location corresponding to the container number is obtained.

[0068] In this step, once the target yard is determined, the container slot layout for each area within the yard is already determined. This means that the target yard's container slot layout database is known, and thus which slots are visible and which are invisible. Furthermore, because the database is based on factors such as container acquisition type, entry and exit times, and shipping and receiving addresses, once the container number of an unrecognizable container is determined, relevant information about the container can be searched based on the number to match it with the most reliable container slot.

[0069] In addition, in some embodiments, after determining the container locations for each container in the target area, the method further includes: allocating container locations for the next new container to enter the container yard based on the determination result, and the allocation method is:

[0070] Construct the overall objective function of container allocation, which aims to minimize the vehicle's passage cost and the number of container unloading in the target yard;

[0071] The following constraints were determined: each slot can only store one container, a container can only occupy one slot, the initial position of the yard crane is at the first bay, the total number of allocated containers is less than the total capacity of the corresponding area, no container can be left hanging in the air, containers are placed in the order of first down and then up, and the number of containers stored in each container area is between the maximum and minimum capacity.

[0072] Under the constraints, the preset algorithm is used to solve the objective function and obtain the allocation result of the new containers entering the yard.

[0073] In some embodiments, the overall objective function is expressed as:

[0074] Q=min(r1×E+r2×F)

[0075] Where Q is the total objective function, r1 is the cost of a vehicle traveling one bay, r2 is the cost of unloading a container, E is the objective function with the minimum number of unloadings, and F is the objective function with the minimum cost of vehicle travel.

[0076] In some embodiments, the objective function for minimizing the number of box flips is:

[0077]

[0078] Where i represents the i-th bay; j represents the j-th row; n represents the n-th container; k represents the k-th layer; I, J, N, K are the number of bays, the number of rows, the total number of containers entering the yard, and the number of layers in the yard, respectively; Lnijk It indicates whether there is a container transfer for the exit container m at the lower position (i,j,kz) when the exit container n is placed at the position (i,j,k). When the container at the position (i,j,k) leaves the yard later than the container at the position (i,j,kz), a container transfer is recorded. The exit probability of a container is related to its attribute value. The attribute value of a container is related to its delivery address, exit time level, weight level, and entry sequence. The larger the attribute value of a container, the heavier its weight and the earlier it leaves the yard.

[0079] In some embodiments, the objective function for minimizing vehicle travel cost is:

[0080]

[0081] Where, P nijk Indicates whether the container location (i, j, k) is stacked with the exit container n; P (n-1)ijk Indicates whether the container location (i, j, k) is stacked with the exit container n-1.

[0082] It should be noted that the preset algorithm is an ant colony algorithm or a genetic algorithm. Under the above constraints, using the ant colony algorithm or the genetic algorithm to solve the total objective function is a common technology and will not be described in detail in this application.

[0083] like Figure 2 、 Figure 3 As shown, an embodiment of the present invention provides a device for determining the container location in a container yard. The device embodiment can be implemented by software, hardware, or a combination of software and hardware. From the hardware level, Figure 2 The figure shows a hardware architecture diagram of a computing device where a container yard location determination device is provided in an embodiment of the present invention. Figure 2 In addition to the processor, memory, network interface, and non-volatile memory shown, the computing device in the embodiment may also include other hardware, such as a forwarding chip responsible for processing messages, etc. Taking software implementation as an example, Figure 3 As shown, as a device in a logical sense, it is formed by the CPU of the computing device in which it is located reading the corresponding computer program in the non-volatile memory into the internal memory and running it.

[0084] Please refer to Figure 3 An embodiment of the present invention provides a device for determining a container location in a container yard, the device comprising:

[0085] The construction unit 300 is used to construct a three-dimensional model of the target container yard; the three-dimensional model includes multiple container areas, each container area includes multiple stacking locations, and each stacking location is used to store containers;

[0086] An acquisition unit 302 is configured to acquire multiple container images of a target area, where the target area is at least a portion of a target container yard;

[0087] A stitching unit 304 is used to stitch each container image to obtain a panoramic image of the target area;

[0088] A determination unit 306 is configured to detect recognizable containers in the panoramic image based on a pre-trained detection model, and to determine whether there are unrecognizable containers in the target area;

[0089] A first determining unit 308 is configured to execute a first solution for the identifiable containers in the target area, wherein the first solution is to determine the location of each identifiable container in the target area based on a mapping relationship between the three-dimensional model and the panoramic image;

[0090] The second determination unit 310 is used to execute a second solution for unrecognizable containers in the target area. The second solution is: based on the entry and exit information of the containers in the target area, the type of container cargo and the shipping and receiving locations, the recognizable container information and the stacking information of the unrecognizable containers in the target area, the second solution is to determine the container location of each unrecognizable container in the target area.

[0091] In some embodiments, the construction unit 300 is configured to perform the following operations:

[0092] Collect geographic coordinate information of each container area in the target yard;

[0093] Based on the geographic coordinate information of each container area, a three-dimensional engine is used to perform three-dimensional modeling of the target yard to obtain a three-dimensional model of the target yard, and the correspondence between the three-dimensional model and the actual geographic coordinates is determined based on the three-dimensional map.

[0094] In some embodiments, each container image is captured by an image acquisition device, which is mounted on a mobile platform. The mobile platform moves along a pre-planned path, and a plurality of acquisition points are provided on the pre-planned path. The image acquisition device captures the corresponding container image at each acquisition point according to a preset angle.

[0095] In some embodiments, determining whether there is an unrecognizable container in the target area is based on a first formula:

[0096] The first formula is: Y = A + CDB

[0097] If Y = 0, it is determined that there are no unrecognizable containers in the target area;

[0098] If Y ≥ 1, it is determined that there are unrecognizable containers in the target area;

[0099] Where Y is the number of unrecognizable containers in the target area; A is the number of containers in the target area before the current image acquisition; C and D are the number of containers entering and leaving the target area between the current image acquisition and the previous image acquisition, respectively; B is the number of recognizable containers in the panoramic image corresponding to the current image acquisition; Y, A, B, C, and D are all natural numbers.

[0100] In some implementations, the first determining unit 308 is configured to perform the following operations:

[0101] Based on the position and angle of the acquisition point where the image acquisition device is located when acquiring the panoramic image, a conversion relationship between the coordinate system of the panoramic image and the geographic coordinate system is established to generate a GeoTIFF file with geographic information;

[0102] Based on the panoramic image and the GeoTIFF file, a preset algorithm is used to perform 3D mesh reconstruction and texture mapping on the panoramic image to obtain the coordinates of each identifiable container in the panoramic image;

[0103] Calculating the actual geographic coordinates of each identifiable container in the panoramic image based on the coordinates of each identifiable container in the panoramic image and the correspondence between the three-dimensional model and the actual geographic coordinates;

[0104] The actual geographic coordinates of each identifiable container in the panoramic image are compared with the container location layout database of the target yard to obtain the location of each identifiable container.

[0105] In some implementations, the second determining unit 310 is configured to perform the following operations:

[0106] Determine a set of stacking locations where images cannot be captured based on a target area's bin layout database and a result of determining the bin locations of the target area before the current image capture.

[0107] Calculating the number of unrecognizable containers in the target area based on the first formula;

[0108] Based on the container number of each container in the target area before the current image acquisition, the container numbers of containers entering and leaving the target area between the current image acquisition and the previous image acquisition, and the container numbers of identifiable containers in the panoramic image corresponding to the current image acquisition, the container number set corresponding to the unidentifiable containers in the target area is determined; the container number of each container in the target area is unique, and the corresponding container's cargo type, entry and exit time, and shipping address are found by the container number;

[0109] Based on the loading and unloading rules of the target area, the cargo type corresponding to each container number, the entry and exit time and the shipping and receiving address, a stacking location is determined for each container number from the stacking location set where images cannot be collected, and the container location corresponding to the container number is obtained.

[0110] In some embodiments, after determining the container locations for each container in the target area, the method further includes: allocating container locations for the next incoming container based on the determination result, using the allocation method as follows:

[0111] Construct the overall objective function of container allocation, which aims to minimize the vehicle's passage cost and the number of container unloading in the target yard;

[0112] The following constraints were determined: each slot can only store one container, a container can only occupy one slot, the initial position of the yard crane is at the first bay, the total number of allocated containers is less than the total capacity of the corresponding area, no container can be left hanging in the air, containers are placed in the order of first down and then up, and the number of containers stored in each container area is between the maximum and minimum capacity.

[0113] Under the constraints, the preset algorithm is used to solve the objective function and obtain the allocation result of the new containers entering the yard.

[0114] It should be noted that the container yard location determination device provided in the above embodiment is merely exemplified by the division of the aforementioned functional modules. In actual applications, the aforementioned functions can be assigned to different functional modules as needed, i.e., the internal structure of the device can be divided into different functional modules to perform all or part of the functions described above. Furthermore, the container yard location determination device provided in the above embodiment and the container yard location determination method embodiment are based on the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0115] The embodiment of the present application also provides a computer device, please refer to Figure 3 The computer device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the container yard location determination method provided by the above-mentioned method embodiments.

[0116] An embodiment of the present application also provides a computer-readable storage medium, on which is stored at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by a processor to implement the container yard location determination method provided by the above-mentioned method embodiments.

[0117] An embodiment of the present application also provides a computer program product, which includes a computer program. A processor of a computer device reads the computer program from a computer-readable storage medium, and the processor executes the computer program, so that the computer device executes the container yard location determination method described in any of the above embodiments.

[0118] For the convenience of description, the above systems or devices are described as being divided into various modules or units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0119] Through the description of the above embodiments, it can be seen that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present application or certain parts of the embodiments.

[0120] Finally, it should be noted that, in this document, relational terms such as first, second, third, and fourth are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

[0121] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for determining container space in a container yard, characterized in that: The method comprises: Constructing a three-dimensional model of a target container yard; the three-dimensional model includes a plurality of container areas, each of the container areas includes a plurality of stacking locations, and each of the stacking locations is used to store containers; Acquire multiple container images of a target area, where the target area is at least a portion of the target yard; Stitching each of the container images to obtain a panoramic image of the target area; Detecting recognizable containers in the panoramic image based on a pre-trained detection model and determining whether there are unrecognizable containers in the target area; if not, executing the first solution; if so, executing the first solution for recognizable containers and executing the second solution for unrecognizable containers; The first solution is: determining the location of each identifiable container based on a mapping relationship between the three-dimensional model and the panoramic image; The second solution is to determine the location of each unrecognizable container based on the entry and exit information of the containers in the target area, the type of container cargo and the shipping and receiving locations, the identifiable container information and the stack location information of the unrecognizable containers; The determination of whether there is an unrecognizable container in the target area is based on the first formula: The first formula is: Y=A+CDB If Y=0, it is determined that there is no unrecognizable container in the target area; If Y ≥ 1, it is determined that there is an unrecognizable container in the target area; Where Y is the number of unrecognizable containers in the target area; A is the number of containers in the target area before the current image acquisition; C and D are the number of containers entering and leaving the target area between the current image acquisition and the previous image acquisition, respectively; B is the number of recognizable containers in the panoramic image corresponding to the current image acquisition; Y, A, B, C, and D are all natural numbers.

2. The method according to claim 1, characterized in that The three-dimensional model of the target storage yard is constructed, including: Collect geographic coordinate information of each container area in the target yard; Based on the geographic coordinate information of each container area, the target yard is three-dimensionally modeled using a three-dimensional engine to obtain a three-dimensional model of the target yard, and the correspondence between the three-dimensional model and the actual geographic coordinates is determined based on a three-dimensional map.

3. The method according to claim 2, characterized in that Each container image is captured by an image acquisition device, which is mounted on a mobile platform. The mobile platform moves along a pre-planned path, and a plurality of acquisition points are provided on the pre-planned path. The image acquisition device captures a corresponding container image at each acquisition point according to a preset angle.

4. The method according to claim 3, characterized in that The determining of the location of each identifiable container based on the mapping relationship between the three-dimensional model and the panoramic image includes: Based on the position and angle of the acquisition point where the image acquisition device is located when acquiring the panoramic image, a conversion relationship between the coordinate system of the panoramic image and the geographic coordinate system is established to generate a GeoTIFF file with geographic information; Based on the panoramic image and the GeoTIFF file, performing three-dimensional mesh reconstruction and texture mapping on the panoramic image based on a preset algorithm to obtain the coordinates of each identifiable container in the panoramic image; Calculating the actual geographic coordinates of each identifiable container in the panoramic image based on the coordinates of each identifiable container in the panoramic image and the correspondence between the three-dimensional model and the actual geographic coordinates; The actual geographical coordinates of each identifiable container in the panoramic image are compared with the container location layout database of the target yard to obtain the container location of each identifiable container.

5. The method according to claim 1, wherein The determining of the location of each unrecognizable container based on the entry and exit information of the containers in the target area, the container cargo type and the shipping and receiving locations, the identifiable container information and the stacking location information of which the image cannot be captured includes: Determining a set of stacking locations for which images cannot be captured based on a bin layout database of the target area and a bin determination result of the target area before the current image capture; Calculating the number of unrecognizable containers in the target area based on the first formula; Based on the container number of each container in the target area before the current image acquisition, the container numbers of containers entering and leaving the target area between the current image acquisition and the previous image acquisition, and the container numbers of identifiable containers in the panoramic image corresponding to the current image acquisition, a set of container numbers corresponding to unidentifiable containers in the target area is determined; the container number of each container in the target area is unique, and the cargo type, entry and exit time, and shipping address of the corresponding container are found using the container number; Based on the loading and unloading rules of the target area, the cargo type corresponding to each container number, the entry and exit time and the shipping and receiving address, a stacking location is determined for each container number from the stacking location set where images cannot be collected, and the container location corresponding to the container number is obtained.

6. The method according to claim 1, characterized in that After determining the container locations for each container in the target area, the method further includes: allocating container locations for the next new container to enter the container yard based on the determination result, and the allocation method is as follows: Constructing an overall objective function for container allocation, wherein the overall objective function aims to minimize the passage cost of vehicles in the target yard and minimize the number of container unloading times; The following constraints were determined: each slot can only store one container, a container can only occupy one slot, the initial position of the yard crane is at the first bay, the total number of allocated containers is less than the total capacity of the corresponding area, no container can be left hanging in the air, containers are placed in the order of first down and then up, and the number of containers stored in each container area is between the maximum and minimum capacity. Under the constraints, a preset algorithm is used to solve the objective function to obtain the allocation result of the new containers entering the yard.

7. A container yard location determination device, characterized in that: The device comprises: A construction unit is used to construct a three-dimensional model of the target container yard; the three-dimensional model includes a plurality of container areas, each of the container areas includes a plurality of stacking locations, and each of the stacking locations is used to store containers; an acquisition unit, configured to acquire a plurality of container images of a target area, wherein the target area is at least a portion of the target yard; a stitching unit, configured to stitch each of the container images to obtain a panoramic image of the target area; a judgment unit, configured to detect recognizable containers in the panoramic image based on a pre-trained detection model, and to judge whether there are unrecognizable containers in the target area; a first determining unit, configured to execute a first solution for the identifiable containers in the target area, wherein the first solution is to determine a location of each identifiable container in the target area based on a mapping relationship between the three-dimensional model and the panoramic image; a second determining unit, configured to execute a second solution for unrecognizable containers in the target area, the second solution being: determining a location of each unrecognizable container in the target area based on entry and exit information of the containers in the target area, container cargo type and shipping and receiving locations, the recognizable container information, and stack location information for which images cannot be captured; The determination of whether there is an unrecognizable container in the target area is based on the first formula: The first formula is: Y=A+CDB If Y=0, it is determined that there is no unrecognizable container in the target area; If Y ≥ 1, it is determined that there is an unrecognizable container in the target area; Where Y is the number of unrecognizable containers in the target area; A is the number of containers in the target area before the current image acquisition; C and D are the number of containers entering and leaving the target area between the current image acquisition and the previous image acquisition, respectively; B is the number of recognizable containers in the panoramic image corresponding to the current image acquisition; Y, A, B, C, and D are all natural numbers.

8. A computer device, characterized in that: The computer device includes a memory and a processor, the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to implement the steps of any one of the methods described in claims 1-6.

9. A computer-readable storage medium, characterized in that The storage medium stores a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 6.

Images