Express delivery piece separation method, device, system and storage medium

Abstract

The present application relates to a method, device, system and storage medium for separating express items. The method comprises: obtaining a video stream image of the express items to be separated; segmenting the video stream image through a trained instance segmentation network model to obtain mask information of each express item to be separated in the video stream image; determining the position information of each express item to be separated according to each mask information; sending a separation instruction carrying each position information to a single-item separation device, and controlling the single-item separation device to separate each express item to be separated according to the separation instruction. The use of this method can improve the efficiency and accuracy of express item separation.

Description

Technical Field

[0001] The present application relates to the field of logistics management technology, and in particular to a method, device, computer equipment and storage medium for separating express items. Background Art

[0002] With the rapid development of the logistics and express delivery industry, the number of orders is increasing day by day. In order to achieve fully automated sorting of express parcels and improve the sorting efficiency and accuracy of express parcels, it is very important to separate the express parcels into individual pieces.

[0003] However, the express delivery single piece separation equipment currently used has the characteristics of low accuracy and poor equipment stability. The single piece separation equipment station must be equipped with a manual station to meet production needs, resulting in high accuracy and labor costs for single piece separation. Summary of the invention

[0004] Based on this, it is necessary to provide a method, device, system and storage medium for separating express items, which can improve the efficiency and accuracy of express item separation, in order to address the above-mentioned technical problems.

[0005] A method for separating express items, the method comprising:

[0006] Obtain the video stream image of the express to be separated;

[0007] Segmenting the video stream image using a trained instance segmentation network model to obtain mask information of each express item to be separated in the video stream image;

[0008] Determine the location information of each express item to be separated according to each mask information;

[0009] A separation instruction carrying each of the position information is sent to the single-piece separation device, and the single-piece separation device is controlled to separate each of the express items to be separated according to the separation instruction.

[0010] In one embodiment, determining the location information of each express item to be separated according to each mask information includes:

[0011] Extract the target contour of each express delivery to be separated according to each mask information to obtain target contour information;

[0012] The center position coordinates and size information of each express item to be separated are determined according to the target contour information to obtain the position information of each express item to be separated.

[0013] In one embodiment, the separation instruction carrying each of the position information is sent to the single-piece separation device, and the single-piece separation device is controlled to separate each of the express items to be separated according to the separation instruction, including:

[0014] A separation instruction carrying each of the position information is sent to a single separation device

[0015] Determining the priority of each of the corresponding separation instructions according to the distance between each of the position information and the single-piece separation equipment;

[0016] Based on the priority level, the single-piece separation equipment is controlled in turn according to each separation instruction to separate the corresponding express items to be separated.

[0017] In one embodiment, obtaining the video stream image of the express to be separated includes:

[0018] When it is determined that there are express items to be separated on the express evacuation equipment according to the acquired photoelectric sensing information, the camera is triggered to collect images to obtain video stream images of the express items to be separated.

[0019] In one embodiment, the training process of the instance segmentation network model includes:

[0020] Acquire express item sample images, and annotate the express item sample images to obtain an annotated sample data set;

[0021] Get the preset instance segmentation network model;

[0022] The instance segmentation network model is trained by the labeled sample data set until convergence to obtain the instance segmentation network model.

[0023] In one embodiment, the method further comprises:

[0024] When it is detected that each of the express items to be separated after separation is not at the target position of the centering device, a prompt instruction is generated;

[0025] The conveying angle of the centering device is adjusted according to the prompt instruction so that the express items to be separated after separation are at the target position of the centering device.

[0026] A single express delivery separation device, the device comprising:

[0027] An acquisition module, used to acquire the video stream image of the express to be separated;

[0028] A segmentation module, used to segment the video stream image by using a trained instance segmentation network model to obtain mask information of each express item to be separated in the video stream image;

[0029] A determination module, used to determine the location information of each express item to be separated according to each mask information;

[0030] The separation module is used to send a separation instruction carrying each of the position information to the single-piece separation device, and control the single-piece separation device to separate each of the express items to be separated according to the separation instruction.

[0031] A single-piece express separation system, the system comprising an image acquisition device, a conveying buffer device, an evacuation device, a single-piece separation device, a centering device and a control host for controlling the image acquisition device and the single-piece separation machine, the control host comprising a memory and a processor, the memory storing a computer program, and the processor implementing the following steps when executing the computer program:

[0032] Obtain the video stream image of the express to be separated;

[0033] Segmenting the video stream image using a trained instance segmentation network model to obtain mask information of each express item to be separated in the video stream image;

[0034] Determine the location information of each express item to be separated according to each mask information;

[0035] A separation instruction carrying each of the position information is sent to the single-piece separation device, and the single-piece separation device is controlled to separate each of the express items to be separated according to the separation instruction.

[0036] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0037] Obtain the video stream image of the express to be separated;

[0038] Segmenting the video stream image using a trained instance segmentation network model to obtain mask information of each express item to be separated in the video stream image;

[0039] Determine the location information of each express item to be separated according to each mask information;

[0040] A separation instruction carrying each of the position information is sent to the single-piece separation device, and the single-piece separation device is controlled to separate each of the express items to be separated according to the separation instruction.

[0041] The above-mentioned express single-piece separation method, device, system and storage medium obtain a video stream image of the express to be separated; use a trained instance segmentation network model to segment the video stream image to obtain mask information of each express to be separated in the video stream image; determine the position information of each express to be separated according to each mask information; receive a separation instruction carrying each position information sent by a programmable logic controller, and control the single-piece separation device to separate each express to be separated according to the separation instruction; accurately determine the position information of each express to be separated based on deep learning and image recognition processing, and control the single-piece separation device to automatically separate each express to be separated according to the generated position information, thereby improving the efficiency and accuracy of express single-piece separation. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is an application environment diagram of a method for separating express items in one embodiment;

[0043] Figure 2 A schematic diagram of a flow chart of a method for separating express items in one embodiment;

[0044] Figure 3 A schematic diagram of a flow chart of a method for separating express items in another embodiment;

[0045] Figure 4 It is a schematic diagram of the flow of a single express delivery separation system in one embodiment;

[0046] Figure 5 A schematic diagram of an application scenario of a method for separating express items in an embodiment;

[0047] Figure 6 It is a structural block diagram of a single express delivery separation device in one embodiment;

[0048] Figure 7 It is a structural block diagram of a single express delivery separation device in another embodiment;

[0049] Figure 8 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0051] The express delivery single piece separation method provided in this application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 and the single-piece separation device 106 through the network. The terminal 102 obtains the video stream image of the express to be separated from the server 104 through the network; the video stream image is segmented by the trained instance segmentation network model to obtain the mask information of each express to be separated in the video stream image; the position information of each express to be separated is determined according to each mask information; a separation instruction carrying each position information is sent to the single-piece separation device 106, and the single-piece separation device is controlled to separate each express to be separated according to the separation instruction. Among them, the terminal 102 can be but is not limited to various personal computers, laptops, smart phones and tablets, and the server 104 can be implemented with an independent server or a server cluster composed of multiple servers. The single-piece separation device 106 is used to separate express.

[0052] In one embodiment, Figure 2 As shown, a method for separating express items is provided, and the method is applied to Figure 1 The terminal in is used as an example to illustrate, including the following steps:

[0053] Step 202, obtaining a video stream image of the express to be separated.

[0054] The video stream image refers to an image of the area covered by the video sensor captured by a video sensor (eg, a camera).

[0055] Specifically, when the sensing device senses that there are express parcels to be separated in the conveying buffer device, the terminal sends a start command to start the conveying buffer device, and starts the belt to convey the express parcels to be separated to the evacuation device; the evacuation device is used to adjust the intervals of the express parcels to be separated in the width direction perpendicular to the conveying direction, so that the express parcels to be separated are separated by a certain distance to avoid overlapping of the express parcels to be separated. When the sensing device senses that the express parcels to be separated pass through the evacuation device, the terminal receives the "sensing the express parcels to be separated passing through the evacuation device" command sent by the sensing device, triggers the image acquisition device to capture the video stream image of the express parcels to be separated, and sends the captured video stream image to the server for storage, and the terminal obtains the captured video stream image from the server in real time. Optionally, the image acquisition device can send the captured video stream image of the express parcels to be separated to the terminal.

[0056] Step 204 , segment the video stream image using the trained instance segmentation network model to obtain mask information of each express item to be separated in the video stream image.

[0057] The instance segmentation network model may include but is not limited to Mask R-CNN, DeepMask, and YOLACT instance segmentation model, etc. The mask is a binary image composed of 0 and 1, and the mask information refers to the pixel value of each pixel.

[0058] The training process of the instance segmentation network model includes: obtaining sample images of express items, labeling the sample images of express items, and obtaining a labeled sample data set; obtaining a preset instance segmentation network model; training the instance segmentation network model through the labeled sample data set until it converges to a set convergence value, and obtaining a trained instance segmentation network model.

[0059] Specifically, the terminal obtains the video stream image of the express to be separated, preprocesses the acquired video stream image, inputs the preprocessed video stream image into the trained instance segmentation network model for segmentation, obtains the feature map corresponding to the preprocessed video stream image, obtains the candidate ROI feature in each feature map, filters out the candidate ROI feature that does not meet the requirements by performing binary classification and regression processing on the candidate ROI feature, classifies, regresses and generates MASK for the obtained ROI feature, and outputs the mask information of each express to be separated in the video stream image of the express to be separated.

[0060] Step 206, determining the location information of each express item to be separated based on each mask information.

[0061] Specifically, the video stream image is input into the trained instance segmentation network model for segmentation, and each express to be separated in the video stream image is separated to obtain the mask information of each express to be separated in the video stream image, that is, the pixel value of each pixel point; the target contour of each express to be separated is extracted according to each mask information to obtain the target contour information; the coordinate calculation function in the terminal OpenCV function is called, and the target contour information is calculated through the coordinate calculation function to obtain the center position coordinates and size information of each express to be separated, and the corresponding position information is determined according to the center position coordinates and size information of each express to be separated.

[0062] Step 208, sending a separation instruction carrying each location information to the single-piece separation device, and controlling the single-piece separation device to separate each express item to be separated according to the separation instruction.

[0063] Specifically, after the terminal obtains the center position coordinates and size information of each express to be separated in the video stream image, the programmable logic controller unit (PLC) in the terminal generates a separation instruction for each express to be separated, and sends the separation instruction to the single-piece separation device; the PLC controls the drive on the single-piece separation device to separate each express to be separated in turn until all express to be separated are separated, and the separated express is transmitted to the express sorting machine through the conveying equipment, so that the express sorting machine sorts the express and delivers each express to the target location.

[0064] In the above-mentioned express single-piece separation method, a video stream image of the express to be separated is obtained; the video stream image is segmented using a trained instance segmentation network model to obtain mask information of each express piece to be separated in the video stream image; the position information of each express piece to be separated is determined according to each mask information; a separation instruction carrying each position information sent by a programmable logic controller is received, and a single-piece separation device is controlled to separate each express piece to be separated according to the separation instruction; the position information of each express to be separated is accurately determined based on deep learning and image recognition processing, and the single-piece separation device is controlled to automatically separate each express piece to be separated according to the generated position information, thereby improving the efficiency and accuracy of express single-piece separation.

[0065] In another embodiment, Figure 3 As shown, a method for separating express items is provided, and the method is applied to Figure 1 The terminal in is used as an example to illustrate, including the following steps:

[0066] Step 302, obtaining a video stream image of the express to be separated.

[0067] Specifically, when it is determined that there are express items to be separated on the express evacuation equipment according to the acquired photoelectric sensing information, the camera is triggered to collect images to obtain a video stream image of the express items to be separated, wherein the photoelectric sensing information is when the express items to be separated are transported to the express evacuation equipment by the upstream belt conveyor, the photoelectric sensor senses the existence of express items and generates photoelectric sensing information (the photoelectric sensing information can be but is not limited to a digital pulse signal); when the photoelectric sensing information is a preset photoelectric sensing information number, it can be determined that there are express items to be separated on the express evacuation equipment, and the PLC controller unit in the terminal triggers the camera to start image collection to obtain a video stream image of the express items to be separated. By detecting whether there are express items to be separated by photoelectric sensors, the express separation efficiency of single-piece separation equipment in different application scenarios can be improved. For example, the amount of express items during shopping festivals is larger than that of ordinary express items. Determining whether there are express items to be separated on the express evacuation equipment based on the highly sensitive photoelectric sensor can avoid the accumulation of express items to be separated, thereby improving the separation efficiency and accuracy of single-piece separation equipment.

[0068] Step 304 , segment the video stream image using the trained instance segmentation network model to obtain mask information of each express item to be separated in the video stream image.

[0069] Step 306: extract the target contour of each express delivery to be separated according to each mask information to obtain the target contour information.

[0070] The target contour refers to the minimum circumscribed rectangle of each express delivery to be separated.

[0071] Specifically, by calling the function in the terminal, the target contour of each express to be separated is extracted according to each mask information, and the minimum bounding rectangle information is obtained through function calculation, including the position coordinates of each vertex of the minimum bounding rectangle.

[0072] Step 308, determining the center position coordinates and size information of each express item to be separated based on the target contour information, and obtaining the position information of each express item to be separated.

[0073] Specifically, the center position coordinates and size information (size information includes the length and width of the express to be separated) of each express to be separated can be determined based on the position coordinates of each vertex of the minimum circumscribed rectangle, that is, the position information of each express to be separated can be obtained.

[0074] Step 310: Send a separation instruction carrying each location information to the single-piece separation device.

[0075] Step 312: Determine the priority of each corresponding separation instruction according to the distance between each position information and the single-piece separation device.

[0076] Among them, the priority of the separation instruction is determined by the relative distance between each express to be separated and the single-piece separation equipment. The smaller the relative distance between the express to be separated and the single-piece separation equipment, the higher the priority.

[0077] Step 314, based on the priority, control the single-piece separation equipment in turn according to each separation instruction to separate the corresponding express items to be separated.

[0078] Specifically, the priority of the separation instruction of each express to be separated is determined according to the relative distance between each position information and the single-piece separation device. According to the priority of the separation instruction, the drive is controlled in sequence, and each express to be separated is separated by controlling the drive.

[0079] Step 316, when it is detected that the express items to be separated after separation are not in the target position of the centering device, a prompt instruction is generated.

[0080] Specifically, when the device to be separated is transported to the centering device, if the corresponding sensor used to detect whether the express is in the center position does not sense that the express is in the center position, the terminal generates a prompt message, which can be displayed on the terminal display screen or played in the form of voice.

[0081] Step 318, adjusting the conveying angle of the centering device according to the prompt instruction, so that the express items to be separated after separation are at the target position of the centering device.

[0082] Specifically, the conveying angle of the inclined roller conveyor in the centering device is adjusted according to the prompt instruction so that the separated express is in a center position, and the separated express is transported to the express sorting device through the conveying device, so that the express sorting device can sort the express; the express is transported in the center through the centering device to avoid squeezing each other and affecting the stability of the single-piece separation device, which facilitates the sorting of the express and improves the sorting efficiency of the sorting equipment.

[0083] In the above-mentioned express single-piece separation method, the video stream image of the express to be separated is obtained by inputting it into a trained instance segmentation network model to segment the video stream image, so as to obtain the mask information of each express to be separated, and the center position coordinates and size information of the minimum circumscribed rectangle of each express to be separated are determined according to the mask information of each express to be separated, that is, the position information of each express to be separated is obtained; the separation instructions carrying each position information are sent to the single-piece separation device, and based on the priority of each separation instruction, the single-piece separation device is controlled in turn to separate the corresponding express to be separated according to each separation instruction to complete the separation of the express, and the separated express is transmitted to the express sorting device in the center through the centering device, thereby improving the sorting efficiency of the express; wherein, the express to be separated is segmented by the trained instance segmentation network model, and the center position coordinates and size information of each express to be separated are determined, thereby improving the efficiency and accuracy of determining the center position coordinates and size information of each express to be separated, thereby improving the separation efficiency and accuracy of the express items and saving labor costs.

[0084] In one embodiment, Figure 4 As shown, a single-piece express separation system is provided, and the single-piece express separation system includes: the system is sequentially provided with a sensing device, a conveying buffer device, an evacuation device, an image acquisition device, a single-piece separation device, a centering device, a conveying device, and a control host for controlling the image acquisition device and the single-piece separation machine according to the transportation direction of the single express, wherein: the sensing device is used to detect whether there are express items to be separated on the evacuation device; the image acquisition device is used to collect video stream images of express items to be separated. The conveying buffer device is used to use a short belt conveyor to realize the cache of express items by independently controlling the belt conveyor drive. Optionally, the conveying buffer device can control the spacing of express items in the length direction by controlling the start and stop time. The evacuation device is composed of multiple rows of conveying mechanisms, and the conveying directions of adjacent conveying mechanisms are at a certain angle, which is used to open the width direction spacing of express items. The single-piece separation device is composed of 7 rows and 4 columns of independently controlled short belt conveying devices, which are controlled by the control host to realize the separation of express items. The centering device is used to make the separated express items in the center position.

[0085] The control host includes a memory and a processor, and the processor implements the following steps when executing the computer program: obtaining a video stream image of the express to be separated; segmenting the video stream image through a trained instance segmentation network model to obtain mask information of each express to be separated in the video stream image; determining the position information of each express to be separated based on the mask information; sending a separation instruction carrying each position information to a single-piece separation device, and controlling the single-piece separation device to separate each express to be separated according to the separation instruction.

[0086] The following is an application scenario of the express single-piece separation method in an embodiment. The terminal includes a PLC control unit. The express to be separated passes through a buffer device, an evacuation device, an image acquisition device, a single-piece separation device, a centering device and a conveying device in sequence before entering the single-piece separation device. Figure 5 As shown, 1 is a transmission and buffering device, 2 is an evacuation device, 3 is an image acquisition device, 4 is a single-piece separation device, 5 is a rack, 6 is a centering device, and 7 is a conveying device.

[0087] When the express items enter the input buffer device, the sensor of the input buffer device detects the express items and starts the belt to convey the express items to be separated to the evacuation device, so that the express items to be separated are separated by a certain distance; when the sensor senses that the express items to be separated pass the evacuation device, the terminal sends an image acquisition instruction to the image acquisition device, so that the image acquisition device acquires the video stream image of the express items to be separated, and sends the acquired video stream image to the server for storage; the terminal obtains the acquired video stream image from the server in real time, and inputs the video stream image into the trained instance segmentation network model for segmentation, so as to obtain the mask information of each express item to be separated in the video stream image; the position information of each express item to be separated is determined according to the mask information; a separation instruction carrying the position information is sent to the single-piece separation device, and the single-piece separation device is controlled to separate each express item to be separated according to the separation instruction; the separated express items are transported to the express sorting device through the conveying device through the centering device. That is, based on deep learning and image recognition processing, the location information of each express to be separated is accurately determined, and the single-piece separation equipment is controlled to automatically separate each express to be separated according to the generated location information, thereby improving the efficiency and accuracy of express piece separation, and thus improving the sorting efficiency of express pieces.

[0088] In the above-mentioned express single-piece separation system, when the express piece enters the input buffer device, the input buffer device sensor detects the express piece, starts the belt to convey the express piece to be separated to the evacuation device, and separates the express pieces to be separated by a certain distance; when the sensor senses that the express piece to be separated passes through the evacuation device, the terminal sends an image acquisition instruction to the image acquisition device, so that the image acquisition device acquires the video stream image of the express piece to be separated, and sends the acquired video stream image to the server for storage; by acquiring the video stream image of the express piece to be separated; using the trained instance segmentation network model to segment the video stream image, and obtain the mask information of each express piece to be separated in the video stream image; determining the position information of each express piece to be separated according to each mask information; receiving the separation instruction carrying each position information sent by the programmable logic controller, and controlling the single-piece separation device to separate each express piece to be separated according to the separation instruction; accurately determining the position information of each express piece to be separated based on deep learning and image recognition processing, and controlling the single-piece separation device to automatically separate each express piece to be separated according to the generated position information, thereby improving the efficiency and accuracy of express single-piece separation.

[0089] For the specific limitations of the express item separation system, please refer to the limitations of the express item separation method mentioned above, which will not be repeated here.

[0090] It should be understood that although Figure 2 , Figure 4 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 2 , Figure 4 At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0091] In one embodiment, Figure 6 As shown, a single express delivery separation device is provided, including: an acquisition module 602, a segmentation module 604, a determination module 606 and a separation module 608, wherein:

[0092] The acquisition module 602 is used to acquire the video stream image of the express to be separated.

[0093] The segmentation module 604 is used to segment the video stream image through the trained instance segmentation network model to obtain the mask information of each express item to be separated in the video stream image.

[0094] The determination module 606 is used to determine the location information of each express item to be separated according to each mask information.

[0095] The separation module 608 is used to send a separation instruction carrying each location information to the single-piece separation device, and control the single-piece separation device to separate each express item to be separated according to the separation instruction.

[0096] In the above-mentioned express single-piece separation device, a video stream image of the express to be separated is obtained; the video stream image is segmented using a trained instance segmentation network model to obtain mask information of each express to be separated in the video stream image; the position information of each express to be separated is determined according to each mask information; a separation instruction carrying each position information sent by a programmable logic controller is received, and the single-piece separation device is controlled to separate each express to be separated according to the separation instruction; the position information of each express to be separated is accurately determined based on deep learning and image recognition processing, and the single-piece separation device is controlled to automatically separate each express to be separated according to the generated position information, thereby improving the efficiency and accuracy of express single-piece separation.

[0097] In another embodiment, Figure 7 As shown, a single express delivery separation device is provided, which includes, in addition to an acquisition module 602, a segmentation module 604, a determination module 606 and a separation module 608, an extraction module 610, a collection module 612, a centering module 614 and a model training module 616, wherein:

[0098] The extraction module 610 extracts the target contour of each express to be separated according to each mask information to obtain the target contour information.

[0099] In one embodiment, the determination module 606 is further used to determine the center position coordinates and size information of each express item to be separated according to the target contour information, and obtain the position information of each express item to be separated.

[0100] In one embodiment, the determination module 606 is further configured to determine the priority of each corresponding separation instruction sent to the single-piece separation device carrying each position information according to the distance between each position information and the single-piece separation device.

[0101] In one embodiment, the separation module 608 is also used to control the single-piece separation device to separate the corresponding express items to be separated according to each separation instruction in turn based on the priority.

[0102] The acquisition module 612 is used to trigger the camera to acquire images when it is determined that there are express items to be separated on the express evacuation equipment based on the acquired photoelectric sensing information, so as to obtain a video stream image of the express items to be separated.

[0103] The centering module 614 is used to generate a prompt instruction when it is detected that the express items to be separated after separation are not at the target position of the centering device; the conveying angle of the centering device is adjusted according to the prompt instruction so that the express items to be separated after separation are at the target position of the centering device.

[0104] The model training module 616 is used to obtain sample images of express items, annotate the sample images of express items, and obtain an annotated sample data set; obtain a preset instance segmentation network model; train the instance segmentation network model through the annotated sample data set until convergence, and obtain an instance segmentation network model.

[0105] In one embodiment, the video stream images of the express to be separated are input into a trained instance segmentation network model to segment the video stream images, and the mask information of each express to be separated is obtained. The center position coordinates and size information of the minimum circumscribed rectangle of each express to be separated are determined according to the mask information of each express to be separated, that is, the position information of each express to be separated is obtained; the separation instructions carrying the position information are sent to the single-piece separation device, and based on the priority of each separation instruction, the single-piece separation device is controlled in turn to separate the corresponding express to be separated according to each separation instruction to complete the separation of the express, and the separated express is transmitted to the express sorting device in the center through the centering device, thereby improving the sorting efficiency of the express; wherein, the express to be separated is segmented by the trained instance segmentation network model, and the center position coordinates and size information of each express to be separated are determined, which improves the efficiency and accuracy of determining the center position coordinates and size information of each express to be separated, thereby improving the separation efficiency and accuracy of the express, saving labor costs.

[0106] For the specific definition of the express item separation device, please refer to the definition of the express item separation method above, which will not be repeated here. Each module in the above-mentioned express item separation device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0107] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 8As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for separating express items is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0108] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0109] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0110] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0111] Obtain the video stream image of the express to be separated;

[0112] The video stream image is segmented through the trained instance segmentation network model to obtain the mask information of each express item to be separated in the video stream image;

[0113] Determine the location information of each express item to be separated according to each mask information;

[0114] A separation instruction carrying each location information is sent to the single-piece separation device, and the single-piece separation device is controlled to separate each express item to be separated according to the separation instruction.

[0115] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0116] Extract the target contour of each express to be separated according to each mask information to obtain the target contour information;

[0117] The center position coordinates and size information of each express to be separated are determined according to the target contour information to obtain the position information of each express to be separated.

[0118] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0119] Separation instructions with location information sent to the single-piece separation device

[0120] Determine the priority of each corresponding separation instruction according to the distance between each position information and the single-piece separation device;

[0121] Based on the priority, the single-piece separation equipment is controlled in turn according to each separation instruction to separate the corresponding express items to be separated.

[0122] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0123] When it is determined that there are express items to be separated on the express evacuation equipment according to the acquired photoelectric sensing information, the camera is triggered to collect images to obtain video stream images of the express items to be separated.

[0124] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0125] Obtain sample images of express items, annotate the sample images of express items, and obtain annotated sample data sets;

[0126] Get the preset instance segmentation network model;

[0127] The instance segmentation network model is trained until convergence by annotating the sample data set to obtain the instance segmentation network model.

[0128] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0129] When the express items to be separated are not located at the target position of the centering device after detection, a prompt instruction is generated;

[0130] Adjust the conveying angle of the centering device according to the prompt instructions so that the express items to be separated are at the target position of the centering device after separation.

[0131] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0132] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0133] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for separating express items, characterized in that: The method comprises: Obtain the video stream image of the express to be separated; Segmenting the video stream image using a trained instance segmentation network model to obtain mask information of each express item to be separated in the video stream image; Determine the location information of each express item to be separated according to each mask information; Sending a separation instruction carrying each of the position information to a single-piece separation device, and controlling the single-piece separation device to separate each of the express items to be separated according to the separation instruction; The controlling the single-piece separation device to separate the express items to be separated according to the separation instruction comprises: Determining the priority of each of the corresponding separation instructions according to the distance between each of the position information and the single-piece separation device; Based on the priority, controlling the single-piece separation device to separate the corresponding express items to be separated in turn according to each separation instruction; Wherein, the method further comprises: When it is detected that each of the express items to be separated after separation is not at the target position of the centering device, a prompt instruction is generated; According to the prompt instruction, the conveying angle of the centering device is adjusted so that the express items to be separated are at the target position of the centering device after separation, so as to avoid mutual squeezing between the express items to be separated; Among them, the express to be separated passes through the conveying buffer device, evacuation equipment, image acquisition equipment, single-piece separation equipment, the centering equipment and conveying equipment in sequence; the conveying buffer device is used to realize the caching of express items by independently controlling the belt conveyor drive, and the conveying buffer device is used to control the length interval of each express item to be separated by controlling the start and stop time; the evacuation equipment is composed of multiple rows of conveying mechanisms, and the conveying directions of adjacent conveying mechanisms are at an angle to open the width direction interval between each express item to be separated, and the single-piece separation equipment is composed of independently controlled short belt conveying equipment, which is controlled by the control host to realize the separation of express items.

2. The method according to claim 1, characterized in that The determining the location information of each express item to be separated according to each mask information includes: Extract the target contour of each express delivery to be separated according to each mask information to obtain target contour information; The center position coordinates and size information of each express item to be separated are determined according to the target contour information to obtain the position information of each express item to be separated.

3. The method according to claim 1, characterized in that The step of obtaining the video stream image of the express to be separated comprises: When it is determined that there are express items to be separated on the express evacuation equipment according to the acquired photoelectric sensing information, the camera is triggered to collect images to obtain video stream images of the express items to be separated.

4. The method according to claim 1, characterized in that The training process of the instance segmentation network model includes: Acquire express item sample images, and annotate the express item sample images to obtain an annotated sample data set; Get the preset instance segmentation network model; The instance segmentation network model is trained by the labeled sample data set until convergence to obtain the instance segmentation network model.

5. A single express delivery separation device, characterized in that: The device comprises: An acquisition module, used to acquire the video stream image of the express to be separated; A segmentation module, used to segment the video stream image by using a trained instance segmentation network model to obtain mask information of each express item to be separated in the video stream image; A determination module, used to determine the location information of each express item to be separated according to each mask information; A separation module, used for sending a separation instruction carrying each of the position information to the single-piece separation device, and controlling the single-piece separation device to separate each of the express items to be separated according to the separation instruction; The determination module is further used to determine the priority of each of the corresponding separation instructions according to the distance between each of the position information and the single-piece separation device; The separation module is also used to control the single-piece separation device to separate the corresponding express items to be separated according to each separation instruction in turn based on the priority; A centering module is used to generate a prompt instruction when it is detected that each of the express items to be separated after separation is not at the target position of the centering device; and adjust the conveying angle of the centering device according to the prompt instruction so that each of the express items to be separated after separation is at the target position of the centering device to avoid mutual squeezing between the express items to be separated; Among them, the express to be separated passes through the conveying buffer device, evacuation equipment, image acquisition equipment, single-piece separation equipment, the centering equipment and conveying equipment in sequence; the conveying buffer device is used to realize the caching of express items by independently controlling the belt conveyor drive, and the conveying buffer device is used to control the length interval of each express item to be separated by controlling the start and stop time; the evacuation equipment is composed of multiple rows of conveying mechanisms, and the conveying directions of adjacent conveying mechanisms are at an angle to open the width direction interval between each express item to be separated, and the single-piece separation equipment is composed of independently controlled short belt conveying equipment, which is controlled by the control host to realize the separation of express items.

6. The device according to claim 5, characterized in that The device also includes: An extraction module, used for extracting the target contour of each express to be separated according to each mask information, and obtaining target contour information; The determination module is also used to determine the center position coordinates and size information of each express to be separated according to the target contour information, and obtain the position information of each express to be separated.

7. The device according to claim 5, characterized in that The device also includes: The model training module is used to obtain sample images of express items, annotate the sample images of express items, and obtain an annotated sample data set; obtain a preset instance segmentation network model; train the instance segmentation network model through the annotated sample data set until convergence, and obtain the instance segmentation network model.

8. The device according to claim 5, characterized in that The device also includes: The acquisition module is used to trigger the camera to acquire images when it is determined that there are express items to be separated on the express evacuation equipment based on the acquired photoelectric sensing information, so as to obtain the video stream image of the express items to be separated.

9. A single express delivery separation system, characterized in that: The system includes a sensing device, an image acquisition device, a conveying and caching device, an evacuation device, a single-piece separation device, a centering device and a control host for controlling the image acquisition device and the single-piece separation device, wherein the control host includes a memory and a processor, and is characterized in that the processor implements the steps of the method described in any one of claims 1 to 4 when executing a computer program.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

Images