Warehousing binding detection system for box stack type tray warehouse

Abstract

The invention relates to a warehousing binding detection system for a box stack type tray warehouse, and the system is characterized in that when a warehouse management module detects that a to-be-warehoused box stack is conveyed to a preset position through a linear conveyor, the linear conveyor is controlled to stop operating through a conveyor control module; a detection instruction is sent to the detection control module, so that the detection control module controls an image acquisition module to acquire image information, controls a code reader to acquire tray information and controls a weight acquisition module to acquire weight information; after information collection is completed, the warehouse management module controls the linear conveyor to continue to operate through the conveyor control module; and the detection control module obtains the category and the number of each cargo box according to the collected information, and if no abnormal working condition exists, the warehouse management module binds the category and the number of the cargo boxes with the tray information. It can be understood that according to the technical scheme, automation of warehousing end operation can be achieved, the labor cost is saved, and the accuracy and efficiency of binding and detection work are improved.

Description

Technical Field

[0001] The invention relates to the technical field of box stack storage, and in particular to a box stack type pallet storage storage binding detection system. Background Art

[0002] In the entire warehouse operation system, the material information input into the warehouse management system when goods are put into storage is the original information for the traceability of the goods. Subsequent operations such as outbound delivery and picking up goods must be verified based on this information. Therefore, the accuracy of the incoming information is related to the accuracy of the warehousing process.

[0003] When entering goods into a traditional pallet warehouse, workers need to use handheld terminals to bind the goods. The general process is: use a handheld terminal to scan the barcode of the goods to obtain the goods information; manually count and enter the quantity of goods using the keyboard of the handheld terminal; manually confirm that the quantity and information of the goods are correct before entering the goods into the warehouse; use the handheld terminal to upload the warehouse information. This process requires manual work, which is costly and inefficient. It is easy to make mistakes in the process of counting and terminal input, especially for box pallets with a large number of goods, it is difficult to achieve both efficiency and accuracy.

[0004] Traditional warehouse inspection generally involves setting up an inspection gantry on the conveyor before warehouse entry, and installing a beam photoelectric sensor on the left and right columns of the gantry to detect whether the box stack is too high or skewed through the blocking signal of the sensor. In this way, the photoelectric sensor has a very small range of action and cannot accurately detect the stack shape. It can also cause false detection due to problems such as straps and wrapping film floating out. Summary of the invention

[0005] In view of this, an object of the present invention is to provide a box-stack pallet warehouse entry binding detection system to solve the problems of low efficiency and insufficient accuracy in the prior art during box-stack entry binding and detection.

[0006] According to a first aspect of an embodiment of the present invention, a box-stack pallet warehouse entry binding detection system is provided, comprising:

[0007] Warehouse management module, conveyor control module, detection control module, linear conveyor, barcode reader, weight acquisition module and image acquisition module;

[0008] The linear conveyor is used to transport the stacks of boxes to be stored in a linear manner;

[0009] The conveyor control module is used to control the operation of the linear conveyor;

[0010] The warehouse management module is used to send a detection instruction to the detection control module and a stop instruction to the conveyor control module to stop the linear conveyor when it detects that the box stack to be stored is transported to the preset position;

[0011] When receiving the detection instruction, the detection control module controls the image acquisition module to acquire image information of the box stack to be stored, controls the barcode reader to acquire pallet information on the pallet of the box stack to be stored, and controls the weight acquisition module to acquire weight information; after the information acquisition is completed, a collection completion instruction is sent to the warehouse management module, so that the warehouse management module controls the linear conveyor to continue to operate through the conveyor control module;

[0012] The detection control module is also used to obtain the type and quantity of each cargo box on the stack of boxes to be stored according to the image information and weight information, and determine whether the stack of boxes to be stored has an abnormal working condition;

[0013] If there is no abnormal operating condition, the warehouse management module binds the type and quantity of the cargo box with the pallet information.

[0014] Preferably, the box-stack pallet warehouse entry binding detection system further includes:

[0015] Gantry; an image acquisition module is arranged on the gantry, and the image acquisition module is five industrial cameras;

[0016] When sending the detection instruction and the stop instruction, the warehouse management module is also used to send a first detection instruction to the detection control module when it is detected that the box stack to be stored is transported to the first preset position; send a second detection instruction to the detection control module when it is detected that the box stack to be stored is transported to the second preset position, and at the same time, send a stop instruction to the conveyor control module to stop the linear conveyor; send a third detection instruction to the detection control module when it is detected that the box stack to be stored is transported to the third preset position;

[0017] When the detection control module receives the first detection instruction, it uses the first industrial camera to collect the front image information of the stack of boxes to be stored; when the second detection instruction is received, it uses the second industrial camera to collect the top image information of the stack of boxes to be stored, uses the third industrial camera to collect the right image information of the stack of boxes to be stored, and uses the fourth industrial camera to collect the left image information of the stack of boxes to be stored, controls the code reader to collect the pallet information on the pallet of the stack of boxes to be stored, controls the weight collection module to collect the weight information, and after the information collection is completed, sends a collection completion instruction to the warehouse management module, so that the warehouse management module controls the linear conveyor to continue to operate through the conveyor control module; when the third detection instruction is received, it uses the fifth industrial camera to collect the rear image information of the stack of boxes to be stored.

[0018] Preferably, if the detection control module determines that the stack of boxes to be stored has an abnormal condition, it sends an abnormal signal to the warehouse management module;

[0019] After receiving the abnormal signal, the warehouse management module sends an abnormal processing instruction to the conveyor control module, so that the linear conveyor conveys the stack of boxes to be stored into the abnormal processing area.

[0020] Preferably, when the detection control module determines whether the box pallet to be stored has an abnormal condition, it includes:

[0021] Performing target detection on the front image information, the top image information, the right image information, the left image information and the rear image information respectively to obtain the target detection result;

[0022] According to the target detection result, it is determined whether the box pallet to be stored is a mixed pallet abnormality, and if so, it is an abnormal working condition;

[0023] According to the target detection result, it is determined whether the cargo boxes on the stack to be stored are beyond the limit range, and if so, it is an abnormal working condition;

[0024] According to the target detection result, it is determined whether the stack of boxes to be stored is crooked or misplaced, and if so, it is an abnormal working condition;

[0025] Determine whether the stacking method of the cargo boxes on the to-be-stored box stack is abnormal according to the target detection result, and if so, it is an abnormal working condition;

[0026] It is determined whether the number of cargo boxes on the to-be-stored box stack is abnormal based on the target detection result and weight information. If so, it is an abnormal operating condition.

[0027] Preferably, when the detection control module determines whether the number of cargo boxes on the stack to be stored is abnormal according to the target detection result and the weight information, it includes:

[0028] deriving the number of containers on each layer from the target detection result of the top image information, and deriving the number of layers of the container stack to be stored from the target detection result of the front image information, and calculating the first container quantity according to the number of containers on each layer and the number of layers of the container stack to be stored;

[0029] Determine the cargo box material category from the target detection result, and read the basic material information corresponding to the cargo box material category from the database, wherein the basic material information includes the weight of a single cargo box and the weight of a pallet;

[0030] The number of second cargo boxes is calculated based on the weight information collected by the control weight collection module, the weight of a single cargo box and the weight of the pallet;

[0031] Compare the quantity in the first container with the quantity in the second container. If they are inconsistent, the quantity in the containers is abnormal.

[0032] Preferably, before the warehouse management module binds the category and quantity of the cargo box with the pallet information, it is also used to search whether there is an identical item of the pallet information in the inventory table. If so, it determines whether the category of the cargo box is consistent with the category recorded in the inventory table. If so, it sends a permission to enter the warehouse instruction to the conveyor control module. If not, it sends an exception handling instruction to the conveyor control module to enable the linear conveyor to send the stack of boxes to be entered into the exception handling area.

[0033] Preferably, if the warehouse management module retrieves an identical item of the pallet information that does not exist in the inventory table, the category and quantity of the cargo box are bound to the pallet information, the category, quantity and pallet information of the cargo box are inserted into the inventory table, and an instruction to allow entry into the warehouse is sent to the conveyor control module.

[0034] Preferably, a left code reader and a right code reader are also provided on the door frame.

[0035] Preferably, the industrial camera is an RGB industrial camera or an RGBD camera.

[0036] Preferably, the gantry includes four pillars and a gantry beam; a first industrial camera is installed on the first pillar, a third industrial camera and a right code reader are installed on the second pillar, a fourth industrial camera and a left code reader are installed on the third pillar, a fifth industrial camera is installed on the fourth pillar, and a second industrial camera is installed on the gantry beam.

[0037] The technical solution provided by the embodiments of the present invention may have the following beneficial effects:

[0038] It can be understood that the technical solution shown in the present invention relates to a box-stack pallet warehouse entry binding detection system. When the warehouse management module detects that the box stack to be entered is transported to the preset position by the linear conveyor, the conveyor control module controls the linear conveyor to stop running, and sends a detection instruction to the detection control module to control the image acquisition module to collect image information, control the code reader to collect pallet information, and control the weight acquisition module to collect weight information; after the information collection is completed, the warehouse management module controls the linear conveyor to continue to run through the conveyor control module; the detection control module derives the category and quantity of each cargo box based on the collected information, and if there is no abnormal working condition, the warehouse management module binds the category and quantity of the cargo box with the pallet information. It can be understood that the technical solution shown in the present invention can realize the automation of the entry-end operation, save labor costs, and improve the accuracy and efficiency of binding and detection work.

[0039] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0041] Figure 1 is a schematic block diagram of a box-stack pallet warehouse entry binding detection system according to an exemplary embodiment;

[0042] Figure 2 is a schematic top view of a detection structure according to an exemplary embodiment;

[0043] Figure 3 is a front view schematic diagram of a detection structure according to an exemplary embodiment;

[0044] Figure 4 is a schematic diagram of a portal frame structure according to an exemplary embodiment;

[0045] Figure 5 is a schematic diagram of a communication interaction process according to an exemplary embodiment;

[0046] Figure 6 is a schematic diagram of an abnormal operating condition according to an exemplary embodiment;

[0047] Figure 7 It is a schematic diagram of an abnormal operating condition detection process according to an exemplary embodiment. DETAILED DESCRIPTION

[0048] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0049] In one embodiment, Figure 1 is a schematic block diagram of a box-type pallet warehouse entry binding detection system according to an exemplary embodiment, see Figure 1 , providing a box-type pallet warehouse entry binding detection system, including:

[0050] Warehouse management module 10, conveyor control module 20, detection control module 30, linear conveyor 40, code reader 70, weight acquisition module 60 and image acquisition module 50;

[0051] The linear conveyor 40 is used to transport the stacks of boxes to be stored in a linear manner;

[0052] The conveyor control module 20 is used to control the operation of the linear conveyor 40;

[0053] The warehouse management module 10 is used to send a detection instruction to the detection control module 30 and send a stop instruction to the conveyor control module 20 to stop the linear conveyor 40 when detecting that the box stack to be stored is transported to the preset position;

[0054] When receiving the detection instruction, the detection control module 30 controls the image acquisition module 50 to acquire image information of the box stack to be stored, controls the barcode reader 70 to acquire pallet information on the box stack to be stored, and controls the weight acquisition module 60 to acquire weight information; after the information acquisition is completed, a collection completion instruction is sent to the warehouse management module 10, so that the warehouse management module 10 controls the linear conveyor 40 to continue to operate through the conveyor control module 20;

[0055] The detection control module 30 is also used to obtain the type and quantity of each cargo box on the stack of boxes to be stored according to the image information and weight information, and determine whether the stack of boxes to be stored has an abnormal working condition;

[0056] If there is no abnormal operating condition, the warehouse management module 10 binds the type and quantity of the cargo box with the pallet information.

[0057] In specific practice, when workers use a box stack type pallet warehouse entry binding detection system shown in this embodiment, they first place the box stack to be entered into the warehouse on the linear conveyor 40. When the linear conveyor 40 transports the box stack to be entered into the warehouse, the warehouse management module 10 can monitor the position of the box stack transportation, and can detect it by setting up photoelectric sensors and other means. When the box stack is transported to the preset position, this position can enable the detection control module 30 to collect corresponding information. The preset position can be multiple, and whether to stop the linear conveyor 40 can be set according to the information to be collected at the preset position. For example, when collecting weight information, the box stack to be entered into the warehouse needs to stop steadily to accurately obtain the weight information. At this time, the linear conveyor 40 can be stopped.

[0058] After the detection control module 30 obtains all the required information, it can determine whether there is any abnormal condition in the box pallet to be stored based on the obtained information. If it is not an abnormal condition, the warehouse management module 10 binds the type and quantity of the cargo box with the pallet information to realize the storage operation.

[0059] It can be understood that the technical solution shown in this embodiment can solve the shortcomings of traditional warehousing binding and detection methods, realize the automation of warehousing operations, save labor costs, and improve the accuracy and efficiency of binding and detection work.

[0060] In another embodiment, see Figure 1 and Figure 2It should be noted that, based on the previous embodiment, the box-stack pallet warehouse entry binding detection system of this embodiment further includes:

[0061] A gantry 80; an image acquisition module 50 is disposed on the gantry 80, and the image acquisition module 50 is five industrial cameras;

[0062] When sending the detection instruction and the stop instruction, the warehouse management module 10 is also used to send the first detection instruction to the detection control module 30 when it is detected that the box stack to be stored is transported to the first preset position; send the second detection instruction to the detection control module 30 when it is detected that the box stack to be stored is transported to the second preset position, and at the same time, send the stop instruction to the conveyor control module 20 to stop the linear conveyor 40; send the third detection instruction to the detection control module 30 when it is detected that the box stack to be stored is transported to the third preset position;

[0063] When receiving the first detection instruction, the detection control module 30 uses the first industrial camera 61 to collect the front image information of the stack of boxes to be stored; when receiving the second detection instruction, the detection control module 30 uses the second industrial camera 62 to collect the top image information of the stack of boxes to be stored, uses the third industrial camera 63 to collect the right image information of the stack of boxes to be stored, and uses the fourth industrial camera 64 to collect the left image information of the stack of boxes to be stored, controls the code reader 70 to collect the pallet information on the pallet of the stack of boxes to be stored, controls the weight collection module 60 to collect the weight information, and after the information collection is completed, sends a collection completion instruction to the warehouse management module 10, so that the warehouse management module 10 controls the linear conveyor 40 to continue to operate through the conveyor control module 20; when receiving the third detection instruction, uses the fifth industrial camera 65 to collect the rear image information of the stack of boxes to be stored.

[0064] In the first aspect, the detection structure provided by this embodiment includes a linear conveyor 40 equipped with a weight acquisition module 60 (such as an electronic scale) and a metal gantry 80 for installing an image acquisition module 50. When the box stack moves on the conveyor, five industrial cameras can respectively capture images of the front, back, left, right, and top surfaces. The code reader 70 scans the barcode or QR code on the left and right sides (or a single side) of the pallet. The camera's photo-taking action is triggered by the occupation of a photoelectric switch at a fixed point on the conveyor. When the photoelectric switch is occupied, it indicates that the box stack to be stored has arrived at the preset position. The warehouse management module 10 sends a corresponding instruction to the detection control module 30, so that the detection control module 30 controls the camera to take pictures. A weight acquisition module 60 is installed on the conveyor for weighing box stacks.

[0065] Secondly, this embodiment provides a detection control module 30, which is a relatively independent functional module for analyzing the multi-angle images taken, the return value of the barcode reader and the return value of the weighing. First, the visual detection algorithm is used to determine whether the box stack to be stored has abnormal conditions such as over-height, skewness, mixed stacking, etc.; then, the normal box stack is identified for each cargo box in its visible surface and the whole is counted.

[0066] The warehouse management module 10 binds the cargo box type and quantity as the material information of the box pallet to be stored with the pallet number scanned by the code reader 70, thereby replacing the traditional manual inventory binding with the handheld terminal.

[0067] Thirdly, this embodiment integrates the warehouse management module 10 and the detection control module 30. As the upper level, the warehouse management module 10 establishes communication with the conveyor control module 20 and the detection control module 30 respectively. During the transmission process, the box stack triggers the photoelectric switch on the conveyor, and the occupancy signal is uploaded to the warehouse management module 10. The warehouse management module 10 sends instructions for taking pictures, reading codes, and weighing to the detection control module 30. The detection control module 30 controls the above sensors to collect signals. After the detection control module 30 calculates the material information of the box stack to be stored, it is uploaded to the warehouse management module to update the inventory table. It can be understood that in this process, the detection control module 30 does not interact directly with the conveyor control module 20. This communication mode facilitates the integration of the detection control module 30 into the existing warehouse management process. During the integrated deployment process, only a small modification needs to be made to the warehouse management module 10, which reduces the difficulty of project implementation.

[0068] This embodiment provides a structural setting method, see Figure 2 , Figure 3 and Figure 4 Preferably, the gantry 80 includes four pillars and a gantry beam 85; the first industrial camera 61 is installed on the first pillar 81, the third industrial camera 63 and the right code reader 82 are installed on the second pillar 82, the fourth industrial camera 64 and the left code reader 71 are installed on the third pillar 83, the fifth industrial camera 65 is installed on the fourth pillar 84, and the second industrial camera 62 is installed on the gantry beam 85. The gantry 80 is also provided with a left code reader 71 and a right code reader 72 for reading the barcode or QR code pasted on both sides of the pallet to obtain the pallet number. According to the pasting situation of the pallet number in the project, the code reader can be installed on only one side of the gantry 80 to save costs.

[0069] Preferably, the industrial camera is an RGB industrial camera or an RGBD camera. According to actual detection conditions, the RGB industrial camera can be replaced with an RGBD camera capable of measuring depth to improve accuracy.

[0070] See also Figure 2 and Figure 3When the box stack is in shooting position A (first preset position), the first industrial camera 61 captures the front image of the box stack. When the box stack is in shooting position B (second preset position), the second industrial camera 62 captures the top image of the box stack, the third industrial camera 63 captures the right image, and the fourth industrial camera 64 captures the left image. The left code reader 71 and the right code reader 72 read the pallet number; the electronic scale installed on the conveyor weighs the box stack. When the box stack is in shooting position C (third preset position), the fifth industrial camera 65 captures the back image of the box stack. The box stack does not stop when passing shooting position A and shooting position C. The camera captures the image during the movement of the box stack, stops when passing shooting position B, and waits for it to stop and be weighed before moving forward.

[0071] See also Figure 5 , showing the communication interaction process of the warehouse management module (warehouse control system WCS), the conveyor control module 20 and the detection control module (Warehousing Binding and Detection System, abbreviated as WBDS).

[0072] First, the upper warehouse management module (WCS) issues a warehousing operation instruction, and the entire system enters the warehousing inspection and inventory mode and waits for the end instruction. The issuance and end of the instruction are realized by the staff manipulating the warehouse management module 10 according to the actual work requirements.

[0073] Then the warehouse management module 10 checks whether there are pallets and goods on the conveyor of the gantry 80 device section. If there are, it sends a command to the conveyor control module 20 to clear the conveyor. A specific implementation method can be: photoelectric sensors are installed at both ends of the warehouse entry detection device section and the three conveyor sections connected to it. When the photoelectric sensors at both ends or the shooting position of the warehouse entry detection device conveyor have an occupied signal, the front conveyor stops waiting; when the current cargo box inspection is completed, the cargo box enters the rear conveyor and the conveyor of the inspection section has no occupied signal, the front conveyor will transport the waiting cargo box to the inspection section. In the whole process, only one pallet of cargo boxes is allowed to appear on the conveyor of the inspection section.

[0074] When the stack of boxes to be stored passes through the shooting position A, the photoelectric switch is triggered. The warehouse management module 10 detects the occupancy signal of the photoelectric switch at position A, and sends a front camera shooting instruction to the detection control module 30. The detection control module 30 receives the instruction to control the first industrial camera 61 to take pictures, and performs target detection on the image in real time, calculates the type and quantity of the boxes in the front view, and temporarily stores the calculation results for post-processing.

[0075] When the box stack to be stored passes through the shooting position B, the photoelectric switch is triggered. The warehouse management module 10 detects the occupancy signal of the photoelectric switch at position B and sends a stop command to the conveyor control module 20. The conveyor stops moving. After the warehouse management module 10 detects that the conveyor has stopped, it sends the left, right, and top camera shooting instructions to the detection control module 30. The detection control module 30 receives the instructions to control the corresponding camera to take pictures, performs target detection on the three images respectively, calculates the types and quantities of the cargo boxes in the left, right, and top views, and temporarily stores the calculation results for post-processing. At this time, the detection control module 30 actively collects the measurement values ​​of the code reader 70 and the electronic scale, and the pallet number and the weight of the box stack are also temporarily stored in the calculation results. The detection control module 30 sends a shooting position B process end signal to the warehouse management module 10, and the warehouse management module 10 sends a command to start the conveyor to the control system and detects whether the conveyor has been started.

[0076] Preferably, at this time, in order to ensure the continuity of the warehousing operation and avoid the accumulation of box stacks at the warehousing end, the detection control module 30 does not return the image detection end signal to the warehouse management module 10. The shooting position B uses the weight data returned by the electronic scale as the end of the detection judgment, and adds a timeout error release function. In this way, the conveyor will not stop moving due to problems with the detection and counting algorithm, but will be used as an exception to transfer the box stack to the warehousing exception processing area.

[0077] The stack of boxes to be stored continues to move forward and triggers the photoelectric switch when passing the shooting position C. The warehouse management module 10 detects the occupancy signal of the photoelectric switch at position C and sends a rear camera shooting instruction to the detection control module 30. The detection control module 30 receives the instruction to control the fifth industrial camera 65 to take a photo, and performs target detection on the image in real time, calculates the type and quantity of the boxes in the rear view, and the calculation result is integrated with the temporarily stored results of the first two shooting positions for post-processing. The post-processing process is described in detail later.

[0078] It should be noted that if the detection control module 30 determines that the stack of boxes to be stored has an abnormal condition, it sends an abnormal signal to the warehouse management module 10; after receiving the abnormal signal, the warehouse management module 10 sends an abnormal processing instruction to the conveyor control module 20, so that the linear conveyor 40 sends the stack of boxes to be stored into the abnormal processing area.

[0079] For abnormal working conditions, the detection control module 30 will return an error signal to the warehouse management module 10, and the warehouse management module 10 will send an abnormal processing operation to the conveyor control module 20. The conveyor will transfer the abnormal box stacks to the abnormal processing area, and the box stacks in the abnormal processing area will be processed manually.

[0080] It should be noted that, for the box stacks whose abnormal conditions are not detected by the detection control module 30, the warehouse management module 10 is also used to retrieve whether there is an identical item of the pallet information in the inventory table. If so, it means that the box stack is returned to the warehouse, and it is determined whether the category of the cargo box is consistent with the category recorded in the inventory table. If so, an instruction to allow entry is sent to the conveyor control module 20. If not, an exception handling instruction is sent to the conveyor control module 20 to enable the linear conveyor 40 to send the box stack to be entered into the exception handling area.

[0081] It should be noted that if the warehouse management module 10 retrieves an identical item of the pallet information that does not exist in the inventory table, it means that the box pallet is a newly entered cargo. The type and quantity of the cargo box are bound to the pallet information, and the type, quantity and pallet information of the cargo box are inserted into the inventory table, and a warehousing permission instruction is sent to the conveyor control module 20.

[0082] The following is an explanation of the detection control module 30 in determining whether the stack of boxes to be stored has an abnormal condition, see Figure 6 Four abnormal operating conditions are shown.

[0083] It should be noted that when the detection control module 30 determines whether the stack of boxes to be stored has an abnormal condition, the determination process is carried out from easy to difficult, including the following steps, see the flowchart Figure 7 :

[0084] Step S11, performing target detection on the front image information, the top image information, the right image information, the left image information and the rear image information respectively to obtain a target detection result.

[0085] In specific practice, a deep learning target detection algorithm can be used to calculate the cargo box detection frame in the images taken at five angles. The target detection network model needs to be trained in advance, with all types of cargo box stacks stored in the warehouse as the data set. In theory, the richer the data set, the better the trained model effect. The specific training process is not the focus of this invention and will not be described here.

[0086] Step S12: judging whether the box pallet to be stored is a mixed pallet abnormality according to the target detection result; if so, it is an abnormal working condition.

[0087] See Mixed Stack Abnormality Figure 6 In the example in the lower right corner, for a box-stack pallet warehouse, one pallet number can only correspond to one material code, that is, only the same type of box can appear on the pallet. Mixing different materials on the pallet will cause management confusion. This anomaly mainly occurs in the inspection of returning to the warehouse after leaving the warehouse.

[0088] The detection method is as follows: first, count the detection box categories in the five surface images to determine whether there are multiple cargo box category IDs. If the number of categories is greater than or equal to 2, it is determined to be a mixed stacking anomaly.

[0089] Step S13: judging whether the cargo boxes on the to-be-stored box stack are beyond the specified range according to the target detection result; if so, it is an abnormal working condition.

[0090] See Beyond Limits Figure 6 In the example in the upper left corner, in order to prevent the cargo boxes from interfering with mechanical structures such as loading platforms, stackers, and shelves when entering the warehouse, it is necessary to limit the overall length, width, and height of the cargo box stacking. Generally, the length and width dimensions do not exceed the range of the pallet, and the height does not exceed 1.5m, which is specific according to the warehouse structure regulations.

[0091] To detect whether the cargo box exceeds the specified range, the relative position relationship between the pallet detection frame and the cargo box detection frame in the front, back, left, and right views is calculated to determine whether the cargo box exceeds the pallet. Since the front and back views are taken during the movement of the pallet, there may be a slight error in the shooting position, so the absolute position of the pixel coordinates of the detection frame in the left and right views can be used to determine whether the pallet is too high.

[0092] Step S14: determine whether the box stack to be stored is crooked or misaligned according to the target detection result. If so, it is an abnormal working condition.

[0093] See also Figure 6 In the example in the upper right corner, stacking and transportation may cause the cargo boxes to be misplaced or the entire stack to be skewed. Even if each cargo box is within the specified range, the offset center of gravity may easily cause it to overturn during warehousing operations, causing risks. Figure 6 The example in the upper right corner shows an abnormal stacking type in which the cargo boxes are skewed. In addition, there may be stacking types in which the cargo boxes are rotated and misaligned, which are not shown in the figure, but should all be judged as abnormal operating conditions.

[0094] When performing skewed and misaligned stack detection, take one of the views as an example: the cargo box detection frame is layered, the pixel coordinates of the left and right boundaries of each layer of the detection frame are calculated, and the results of the bottom layer are used to compare the results of other layers. If the difference exceeds the set threshold, it is determined to be a skewed stack.

[0095] Step S15: judging whether the stacking mode of the cargo boxes on the to-be-stored box stack is abnormal according to the target detection result; if so, it is an abnormal working condition.

[0096] For abnormal stacking mode, see Figure 6 In the example in the lower left corner, for the boxes in the box stack, the stacking should comply with certain rules. The boxes are allowed to rotate around the Z axis in units of 90 degrees, but are not allowed to rotate around the X or Y axis to appear sideways.

[0097] The stacking method error is the most difficult to judge. For cargo boxes with large differences in length, width and height, the aspect ratio of each detection frame can be compared for judgment, but there may be a misjudgment caused by a large change in the aspect ratio of the detection frame after the cargo box is rotated 90 degrees around the Z axis. Therefore, the most reliable solution is to judge by the surface texture of the cargo box. For different types of cargo boxes, a universal direction identifier can be printed on the outer packaging and a separate target detection network can be trained to judge the cargo box stacking method.

[0098] Step S16: determine whether the number of cargo boxes on the stack to be stored is abnormal based on the target detection result and weight information. If so, it is an abnormal operating condition.

[0099] After the above four abnormality judgments are completed, counting verification is performed.

[0100] It should be noted that the number of containers on each layer is obtained from the target detection result of the top image information, the number of layers of the stacks of containers to be stored is obtained from the target detection result of the front image information, and the first container quantity is calculated based on the number of containers on each layer and the number of layers of the stacks of containers to be stored (the first container quantity is equal to the number of layers multiplied by the number of each layer).

[0101] Determine the cargo box material category from the target detection result, and read the basic material information corresponding to the cargo box material category from the database, wherein the basic material information includes the weight of a single cargo box and the weight of a pallet;

[0102] The second container quantity is calculated based on the weight information collected by the control weight collection module 60, the weight of a single container and the weight of the pallet (the total weight of the stack obtained by weighing minus the weight of the pallet is divided by the weight of a single container to obtain the second container quantity in the stack to be stored).

[0103] Compare the quantity in the first container with the quantity in the second container. If they are inconsistent, the quantity in the containers is abnormal.

[0104] The material type and quantity of the pallets to be stored are written into the temporary storage table as material information together with the pallet number. This step completes the storage binding. The subsequent warehouse management module 10 will update the inventory table according to the temporary storage table to realize automated storage operations.

[0105] It can be understood that the detection structure involved in the present invention is used to fix the camera and the code reader, and is a hardware structure for automated warehousing binding and detection. A detection control module WBDS is proposed and designed to realize functions such as vision-based abnormal condition detection of box stacks, box counting, and material pallet binding. An integration solution of a warehouse control system WCS and WBDS is provided, so that WBDS can be integrated and docked with the existing warehouse control system as a functional module.

[0106] It realizes automatic storage binding without manual operation. It uses cameras to automatically detect and count box stacks instead of manual visual inspection, which reduces labor costs and improves inspection efficiency. WBDS can be called by WCS as an independent functional module, and is easy to integrate with the existing warehouse scheduling system.

[0107] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0108] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0109] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0110] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0111] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0112] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0113] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0114] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0115] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A box-stack pallet warehouse entry binding detection system, characterized in that: include: Warehouse management module, conveyor control module, detection control module, linear conveyor, barcode reader, weight acquisition module and image acquisition module; The linear conveyor is used to transport the stacks of boxes to be stored in a linear manner; The conveyor control module is used to control the operation of the linear conveyor; The warehouse management module is used to send a detection instruction to the detection control module and a stop instruction to the conveyor control module to stop the linear conveyor when it detects that the box stack to be stored is transported to the preset position; When receiving the detection instruction, the detection control module controls the image acquisition module to acquire image information of the box stack to be stored, controls the barcode reader to acquire pallet information on the pallet of the box stack to be stored, and controls the weight acquisition module to acquire weight information; after the information acquisition is completed, a collection completion instruction is sent to the warehouse management module, so that the warehouse management module controls the linear conveyor to continue to operate through the conveyor control module; The detection control module is also used to obtain the type and quantity of each cargo box on the stack of boxes to be stored according to the image information and weight information, and determine whether the stack of boxes to be stored has an abnormal working condition; If there is no abnormal operating condition, the warehouse management module binds the type and quantity of the cargo box with the pallet information.

2. The box-stack pallet warehouse entry binding detection system according to claim 1 is characterized in that: Also includes: Gantry; an image acquisition module is arranged on the gantry, and the image acquisition module is five industrial cameras; When sending the detection instruction and the stop instruction, the warehouse management module is also used to send a first detection instruction to the detection control module when it is detected that the box stack to be stored is transported to the first preset position; send a second detection instruction to the detection control module when it is detected that the box stack to be stored is transported to the second preset position, and at the same time, send a stop instruction to the conveyor control module to stop the linear conveyor; send a third detection instruction to the detection control module when it is detected that the box stack to be stored is transported to the third preset position; When the detection control module receives the first detection instruction, it uses the first industrial camera to collect the front image information of the stack of boxes to be stored; when the second detection instruction is received, it uses the second industrial camera to collect the top image information of the stack of boxes to be stored, uses the third industrial camera to collect the right image information of the stack of boxes to be stored, and uses the fourth industrial camera to collect the left image information of the stack of boxes to be stored, controls the code reader to collect the pallet information on the pallet of the stack of boxes to be stored, controls the weight collection module to collect the weight information, and after the information collection is completed, sends a collection completion instruction to the warehouse management module, so that the warehouse management module controls the linear conveyor to continue to operate through the conveyor control module; when the third detection instruction is received, it uses the fifth industrial camera to collect the rear image information of the stack of boxes to be stored.

3. The box-stack pallet warehouse entry binding detection system according to claim 2 is characterized in that: If the detection control module determines that the stack of boxes to be stored has an abnormal condition, it sends an abnormal signal to the warehouse management module; After receiving the abnormal signal, the warehouse management module sends an abnormal processing instruction to the conveyor control module, so that the linear conveyor conveys the stack of boxes to be stored into the abnormal processing area.

4. The box-stack pallet warehouse entry binding detection system according to claim 2 is characterized in that: When the detection control module determines whether the stack of boxes to be stored has an abnormal condition, it includes: Performing target detection on the front image information, the top image information, the right image information, the left image information and the rear image information respectively to obtain the target detection result; According to the target detection result, it is determined whether the box pallet to be stored is a mixed pallet abnormality, and if so, it is an abnormal working condition; According to the target detection result, it is determined whether the cargo boxes on the stack to be stored are beyond the limit range, and if so, it is an abnormal working condition; According to the target detection result, it is determined whether the stack of boxes to be stored is crooked or misplaced, and if so, it is an abnormal working condition; Determine whether the stacking method of the cargo boxes on the to-be-stored box stack is abnormal according to the target detection result, and if so, it is an abnormal working condition; It is determined whether the number of cargo boxes on the to-be-stored box stack is abnormal based on the target detection result and weight information. If so, it is an abnormal operating condition.

5. The box-stack pallet warehouse entry binding detection system according to claim 4 is characterized in that: When the detection control module determines whether the number of cargo boxes on the stack to be stored is abnormal according to the target detection result and the weight information, it includes: deriving the number of containers on each layer from the target detection result of the top image information, and deriving the number of layers of the container stack to be stored from the target detection result of the front image information, and calculating the first container quantity according to the number of containers on each layer and the number of layers of the container stack to be stored; Determine the cargo box material category from the target detection result, and read the basic material information corresponding to the cargo box material category from the database, wherein the basic material information includes the weight of a single cargo box and the weight of a pallet; The number of second cargo boxes is calculated based on the weight information collected by the control weight collection module, the weight of a single cargo box and the weight of the pallet; Compare the quantity in the first container with the quantity in the second container. If they are inconsistent, the quantity in the containers is abnormal.

6. The box-stack pallet warehouse entry binding detection system according to claim 1 is characterized in that: Before the warehouse management module binds the category and quantity of the cargo box with the pallet information, it is also used to search whether there is the same item of the pallet information in the inventory table. If so, it determines whether the category of the cargo box is consistent with the category recorded in the inventory table. If so, it sends a permission to enter the warehouse instruction to the conveyor control module. If not, it sends an exception handling instruction to the conveyor control module to enable the linear conveyor to send the box stack to be entered into the exception handling area.

7. The box-stack pallet warehouse entry binding detection system according to claim 6 is characterized in that: If the warehouse management module retrieves an identical item of the pallet information that does not exist in the inventory table, it binds the category and quantity of the cargo box with the pallet information, inserts the category, quantity and pallet information of the cargo box into the inventory table, and sends a warehousing permission instruction to the conveyor control module.

8. The box-stack pallet warehouse entry binding detection system according to claim 2 is characterized in that: The door frame is also provided with a left code reader and a right code reader.

9. The box-stack pallet warehouse entry binding detection system according to claim 2 is characterized in that: The industrial camera is an RGB industrial camera or an RGBD camera.

10. The box-stack pallet warehouse entry binding detection system according to claim 8, characterized in that: The gantry includes four pillars and a gantry beam; the first industrial camera is installed on the first pillar, the third industrial camera and the right code reader are installed on the second pillar, the fourth industrial camera and the left code reader are installed on the third pillar, the fifth industrial camera is installed on the fourth pillar, and the second industrial camera is installed on the gantry beam.

Images