Tray state detection method for tilting tray type sorting machine

Abstract

The invention provides a tray state detection method for a tilting tray type sorting machine. The tray state detection method comprises the following steps that firstly, image information of a tray area is collected through a visual system; 2, analyzing the acquired image information by using a deep learning technology and an image processing technology, identifying a tilting-pan type trolley, a package and a side plate, obtaining an external rectangular frame, obtaining a contour mask image of each target detection object, and extracting contour points of the contour mask image; step 3, carrying out existence detection and double-piece and stacked-piece detection; 4, establishing a coordinate system of the tray trolley and a coordinate system of the running direction, and performing torsion detection on the tray trolley; step 5, performing super-region detection; sixthly, side plate blocking detection is carried out; according to the technical scheme, the tilting-pan type trolley, the parcels and the side plates are recognized, and the external rectangular frame and the contour mask image are obtained, so that detection is more comprehensive, the problem that a traditional detection device is low in recognition accuracy is solved, and the sorting accuracy and efficiency are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of material sorting, and more specifically, particularly relates to a method for detecting the state of a tray of a tilting tray sorting machine. Background Art

[0002] In today's era of rapid development of the logistics industry, the efficiency and accuracy of parcel sorting technology are very important; at present, as a commonly used logistics sorting equipment, the flip tray sorting machine has been widely used in major logistics centers and warehousing systems due to its high efficiency and stable performance.

[0003] However, in actual application, traditional detection devices are easily affected by light and pallet dirt, resulting in low recognition accuracy and failure to promptly detect abnormal situations in the sorting process. Secondly, due to the lack of detection capabilities for abnormal situations such as multiple pieces, overlapping pieces, and out-of-area items, it is easy to cause packages to get stuck, and the pallet trolley is prone to deflection during use, affecting sorting efficiency. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a method for detecting the status of a tray of a flip-tray sorting machine, so as to solve the technical problems in the prior art that the traditional detection device is easily affected by light and tray dirt, resulting in low recognition accuracy; lacks detection of abnormal situations such as multiple pieces, overlapping pieces, out-of-area, and tray trolley twisting, and is prone to package jamming, affecting sorting efficiency.

[0005] The purpose and effect of the tray status detection method of the tilting tray sorting machine of the present invention are achieved by the following specific technical means:

[0006] A method for detecting the state of a tray of a tilting tray sorting machine comprises the following steps:

[0007] Step 1: Collect image information of the pallet area through the visual system;

[0008] Step 2: Analyze the collected image information using deep learning technology and image processing technology, identify the tipping trolley, package and side panel, and obtain their circumscribed rectangular frames, and simultaneously obtain the contour mask image of each target detection object and extract its contour points;

[0009] Step 3: Conduct presence detection and double piece and overlap detection;

[0010] By analyzing the image information, it is determined whether there are packages on the pallet, and then double and overlapping packages are detected. When there are more than two packages on the pallet area, an alarm is issued;

[0011] Step 4: Establish the coordinate system of the pallet trolley and the coordinate system of the running direction, and perform torsion detection of the pallet trolley;

[0012] Step 5: Super-region detection;

[0013] By identifying the location of the package and the pallet, calculating the deviation between the package and the pallet, determining whether the package exceeds the pallet area, and further calculating the specific type of over-area;

[0014] Step 6: Side panel jam detection;

[0015] By identifying the positions of the package and the side panel, the intersection area between the package and the side panel is calculated to determine whether the side panel is stuck.

[0016] In a preferred embodiment, in step 2, the image processing technology includes correcting the image information by Zhang Zhengyou method, rotating the image, and correcting the rotational offset of the image information.

[0017] In a preferred embodiment, in step 4, the pallet trolley torsion detection includes:

[0018] The angle between the pallet trolley coordinate system and the running direction coordinate system is calculated, and when the angle is greater than a set threshold, it is determined that the pallet trolley is twisted.

[0019] In a preferred embodiment, in step 5, the calculation formula of the deviation between the package and the pallet is:

[0020]

[0021] Where S1 represents the ratio of the minimum bounding rectangle area of ​​the package and the pallet to the minimum bounding rectangle area of ​​the package, A represents the area occupied by the minimum bounding rectangle of the package, and B represents the area occupied by the minimum bounding rectangle of the pallet. When S1 is less than the preset threshold of the user, the package is judged to be out of area.

[0022] In a preferred embodiment, the calculation of the super-area type includes calculating a distance L, and the distance L represents the deviation distance between the package and the pallet on the upper side. When the distance L is greater than a set threshold, the package status is judged to be IBB, and the IBB status indicates that the package is stuck between two pallet carts.

[0023] In a preferred embodiment, the calculation of the super-area type also includes calculating the distance K, and the distance K represents the deviation distance between the package and the pallet on the left side. When the distance K is greater than a set threshold, the package status is judged to be IOB, and the IOB status indicates that the package exceeds the edge of the pallet trolley.

[0024] In a preferred embodiment, in step six, the calculation formula for calculating the intersection area between the package and the side panel is:

[0025]

[0026] In the formula, S2 represents the ratio of the minimum circumscribed rectangle area of ​​the side panel, A represents the area occupied by the minimum circumscribed rectangle of the package, and C represents the area occupied by the minimum circumscribed rectangle of the side panel; when S2 is greater than the preset threshold, the package is determined to be stuck in the side panel.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] 1. This method adopts deep learning technology and image processing technology. By correcting and rotating the image information, the quality and accuracy of the image are improved, and the interference caused by light and dirty pallets can be effectively overcome. At the same time, the tipping trolley, parcel and side panel are identified, and their external rectangular frame and contour mask images are obtained, making the detection more precise and comprehensive, solving the problem of low recognition accuracy of traditional detection devices, and being able to timely detect the status of the parcel on the pallet, whether there is a parcel, double or overlapping, it can accurately judge and alarm, improving the accuracy and efficiency of sorting.

[0029] 2. Through the out-of-area detection of the package, it can determine whether the package exceeds the pallet area, and further calculate the specific out-of-area type, and determine whether the package status is IBB package stuck between two pallet trolleys, or IOB package exceeds the edge of the pallet trolley, so as to deal with the out-of-area situation in time; through the side panel jam detection, by calculating the intersection area between the package and the side panel, it can be determined whether the side panel is jammed to ensure the smooth progress of the sorting process; through the pallet trolley torsion detection, the pallet trolley failure can be discovered in time to prevent the equipment from causing greater damage and improve stability; these comprehensive detection functions effectively reduce the failure rate of package jams, improve sorting efficiency, and bring higher benefits and reliability to the logistics sorting industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 The present invention is a flowchart of the steps of a method for detecting the state of a tray of a tray-turning sorting machine.

[0031] Figure 2 It is a schematic diagram of a detection frame obtained by deep learning technology in a tray status detection method for a flip-tray sorting machine of the present invention.

[0032] Figure 3 It is a schematic diagram of the super-area distance K and the distance L in a method for detecting the state of a tray of a tilting tray sorting machine of the present invention.

[0033] Figure 4 It is a schematic diagram of a coordinate system for torsion detection in a method for detecting a tray state of a tilt-tray sorting machine according to the present invention.

[0034] Figure 5It is a state diagram of operation in a method for detecting the state of a tray of a tilting tray sorting machine according to the present invention.

[0035] In the figure, the corresponding relationship between the component names and the figure numbers is as follows:

[0036] 10. Pallet; 11. Side panel; 12. Package. DETAILED DESCRIPTION

[0037] In order to further understand the present invention, preferred embodiments of the present invention are described below in conjunction with the accompanying drawings and examples. However, it should be understood that these descriptions are only for further illustrating the features and advantages of the present invention, rather than limiting the claims of the present invention.

[0038] Example:

[0039] Please refer to Figures 1 to 5 As shown, the present invention provides a method for detecting the state of a tray of a tilting tray sorting machine, comprising the following steps:

[0040] Step 1: The visual inspection device collects image information of the pallet area and takes pictures of the pallet area with preset parameters to ensure that complete and clear image information is obtained, which can provide a reliable image data basis for subsequent inspection steps;

[0041] Step 2: Analyze the collected image information using deep learning technology and powerful image processing technology. Through the deep learning algorithm, identify the tipping trolley, package 12 and side panel 11 in the image, and accurately distinguish different target objects. During the identification process, determine an external rectangular frame for each target object to clarify its position and range in the image. At the same time, obtain the contour mask image of each target object. This process can more clearly outline the contour shape of the target object. Then, extract its contour points from the contour mask image, making the detection more precise and comprehensive, solving the problem of low recognition accuracy of traditional detection devices, and being able to timely detect the status of the package 12 on the pallet 10. These contour points will provide key data support for subsequent analysis and detection, making the feature extraction of the target object more accurate and comprehensive, and improving the accuracy and efficiency of sorting.

[0042] Step 3: Perform presence detection and double or overlapping detection. First, determine whether there is a package 12 on the pallet 10 by analyzing the collected image information; then, perform double or overlapping detection on the packages 12 on the pallet 10. During the detection process, continuously monitor the number and status of the packages 12 in the pallet area. Once more than two packages 12 are found in the pallet area, the alarm system is immediately triggered; the alarm system will sound an alarm in the form of sound and light to remind the staff to deal with abnormal situations in time, ensure the accurate sorting work, and avoid sorting errors and reduced efficiency due to double or overlapping problems;

[0043] Step 4: Establish the coordinate system of the pallet trolley and the coordinate system of the running direction, and perform the torsion detection of the pallet trolley; First, establish the coordinate system of the pallet trolley, which can describe the position and posture of the pallet trolley in space. At the same time, according to the running direction of the sorter, establish the corresponding running direction coordinate system, so as to better analyze the motion state of the pallet trolley; During the detection process, constantly compare the relationship between the coordinate system of the pallet trolley and the running direction coordinate system, and determine whether the pallet trolley is twisted by calculating the angle difference between the two; If the pallet trolley is detected to be twisted, an alarm can be issued in time to remind the staff to make adjustments and repairs to ensure the normal operation of the pallet trolley and the efficient sorting work;

[0044] Step 5: Super-region detection;

[0045] Identify the positions of the parcel 12 and the pallet 10, and determine the specific coordinates of the parcel 12 and the pallet 10 in space; then, calculate the deviation between the parcel 12 and the pallet 10 to determine whether the parcel exceeds the pallet area. Once there is a sign that the parcel 12 exceeds the pallet area, immediately conduct an in-depth analysis and further calculate the specific type of over-area. By evaluating the direction and degree of the parcel 12 exceeding the pallet area, determine whether it belongs to a specific over-area type where the parcel 12 is stuck between two pallet trolleys, or another type of over-area where the parcel 12 exceeds the edge of the pallet trolley, so as to take targeted measures to handle it and ensure the smooth progress of the sorting process;

[0046] Step 6: Side panel jam detection;

[0047] Identify the specific positions of the parcel 12 and the side panel 11 in the sorting machine, then calculate the intersection area between the parcel 12 and the side panel 11, and then determine whether the side panel is jammed based on the calculated intersection area parameters. Compare the preset standard value with the actual calculated value. Once it is found that the intersection area of ​​the parcel 12 and the side panel 11 exceeds a certain range or presents a specific abnormal state, it is determined that there may be a side panel jam, so that corresponding treatment measures can be taken in time to ensure the normal operation of the sorting machine.

[0048] In step two, the image processing technology includes performing a comprehensive correction operation on the image information through the Zhang Zhengyou method. As a mature and effective image correction technology, the Zhang Zhengyou method can accurately analyze the characteristics and parameters of the image and accurately correct various deformations and deviations in the collected image information; specifically, the image is rotated, and the image is adjusted to the correct orientation by calculating the rotation angle and direction of the image to ensure that the target object in the image presents an accurate posture and position relationship; at the same time, the rotation offset of the image information is corrected, and targeted adjustments and repairs are made for possible rotation offset problems caused by shooting angles, equipment errors, etc., so that the image information is more accurate and stable; by correcting and rotating the image information, the quality and accuracy of the image are improved, and the interference caused by light and pallet dirt can be effectively overcome, providing a high-quality image foundation for the subsequent recognition of the tipping trolley, package 12 and side panel 11 and other detection steps.

[0049] In step 4, the pallet trolley torsion detection includes: first, determining the pallet trolley coordinate system and the running direction coordinate system, establishing a coordinate system that can describe the position and posture of the pallet trolley in space, and at the same time, according to the running direction of the sorting machine, establishing a corresponding running direction coordinate system, the coordinate system can reflect the overall movement trend and direction of the sorting machine; then, calculating the angle between the pallet trolley coordinate system and the running direction coordinate system, and calculating the angle difference between them; when the calculated angle is greater than the set threshold, it is determined that the pallet trolley is twisted; in logistics sorting operations, the normal operation of the pallet trolley is very important. Once a fault such as torsion occurs, if it cannot be discovered and handled in time, it may cause greater damage to the equipment and even affect the stability of the entire sorting system; and through this detection method, it can be identified and warned in time when a fault occurs, thereby improving the stability of the equipment.

[0050] In step 5, the calculation formula for the deviation between the package and the pallet is:

[0051]

[0052] In the formula, S1 represents the ratio of the minimum circumscribed rectangle area of ​​the package and the pallet to the minimum circumscribed rectangle area of ​​the package; A represents the area occupied by the minimum circumscribed rectangle of the package, which can reflect the range of the package in space; B represents the area occupied by the minimum circumscribed rectangle of the pallet, which can clearly define the spatial range of the pallet; by calculating the proportional relationship between these two areas, the relative position relationship between the package 12 and the pallet 10 can be effectively determined. When this S1 is less than the preset threshold of the user, the package is determined to be out of area.

[0053] Calculating the super-region type includes calculating the distance L, where the distance L represents the deviation distance between the package 12 and the pallet 10 on the upper side. When the distance L is greater than the set threshold, the package status is determined to be IBB; the IBB status indicates that the package 12 is stuck between two pallet trolleys, which will have an adverse effect on the sorting work; by accurately calculating the distance L and making corresponding status judgments, this abnormal situation can be discovered in time so that effective measures can be taken to deal with it and ensure the smooth progress of the sorting process.

[0054] Calculating the super-region type also includes calculating the distance K, which represents the deviation distance between the package 12 and the pallet 10 on the left side, and detecting the specific excess distance of the package 12 on the left side of the pallet 10. When this distance K is greater than the set threshold, the package status is judged to be IOB; wherein, the IOB status indicates that the package 12 exceeds the edge of the pallet trolley. In actual sorting operations, this situation may cause the package 12 to be unstable during the transmission process, and may even fall or affect the normal sorting of other packages 12; by calculating the distance K and judging the package status based on the comparison result with the set threshold, the situation where the package 12 exceeds the edge of the pallet trolley can be discovered in a timely and effective manner, thereby taking corresponding countermeasures to ensure the stable progress of the sorting work.

[0055] By judging whether the package 12 exceeds the pallet area, the specific out-of-area type is further calculated on this basis. When the package status is judged to be I BB, that is, the package 12 is stuck between the two pallet carts, or when the status of the package 12 is judged to be I OB, that is, the package 12 exceeds the edge of the pallet cart, corresponding processing measures can be taken in time to efficiently deal with the out-of-area situation, thereby avoiding sorting interruptions or errors caused by the package 12 exceeding the area.

[0056] In step six, the calculation formula for calculating the intersection area between the package 12 and the side panel 11 is:

[0057]

[0058] In the formula, S2 represents the ratio of the minimum circumscribed rectangle area of ​​the side panel, A represents the area occupied by the minimum circumscribed rectangle of the package, and C represents the area occupied by the minimum circumscribed rectangle of the side panel, which clearly defines the range of the side panel in space. Through this calculation formula, the size of the intersection area between the package 12 and the side panel 11 can be effectively determined. When this S2 is greater than the preset threshold, the package is determined to be stuck in the side panel. In the logistics sorting process, through this calculation formula, it can be identified whether the package 12 has an abnormal contact with the side panel 11, ensuring that the problem of the package and the side panel being stuck can be discovered and handled in time during the sorting process to prevent the sorting efficiency and the normal operation of the equipment from being affected. Through this detection method, the problem can be discovered and handled in time at an early stage, reducing unnecessary downtime and maintenance time caused by the side panel being stuck.

[0059] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for detecting the status of a tray of a tilting tray sorting machine, characterized in that: The following steps are included: Step 1: Collect image information of the pallet area through the visual system; Step 2: Analyze the collected image information using deep learning technology and image processing technology, identify the tipping trolley, the package (12) and the side panel (11), and obtain their circumscribed rectangular frames, and simultaneously obtain the contour mask image of each target detection object and extract its contour points; Step 3: Conduct presence detection and double piece and overlap detection; By analyzing the image information, it is determined whether there is a package (12) on the pallet (10), and then double and overlapping detection is performed on it, and an alarm is triggered when there is more than one package (12) on the pallet (10); Step 4: Establish the coordinate system of the pallet trolley and the coordinate system of the running direction, and perform torsion detection of the pallet trolley; Step 5: Super-region detection; By identifying the positions of the package (12) and the pallet (10), calculating the deviation between the package (12) and the pallet (10), determining whether the package exceeds the pallet area, and further calculating the specific over-area type; Step 6: Side panel jam detection; By identifying the positions of the package (12) and the side plate (11), the intersection area between the package (12) and the side plate (11) is calculated to determine whether the side plate is stuck.

2. A method for detecting the state of a tray of a tilting tray sorting machine according to claim 1, characterized in that: In step 2, the image processing technology includes correcting the image information through the Zhang Zhengyou method, rotating the image, and correcting the rotational offset of the image information.

3. A method for detecting the state of a tray of a tilting tray sorting machine according to claim 1, characterized in that: In step 4, the pallet trolley torsion detection includes: The angle between the pallet trolley coordinate system and the running direction coordinate system is calculated, and when the angle is greater than a set threshold, it is determined that the pallet trolley is twisted.

4. A method for detecting the state of a tray of a tilting tray sorting machine according to claim 1, characterized in that: In step 5, the calculation formula for the deviation between the package (12) and the pallet (10) is: In the formula, S1 represents the ratio of the minimum circumscribed rectangular area of ​​the package (12) and the pallet (10) to the minimum circumscribed rectangular area of ​​the package, A represents the area occupied by the minimum circumscribed rectangle of the package, and B represents the area occupied by the minimum circumscribed rectangle of the pallet; when S1 is less than the preset threshold of the user, the package is determined to be out of area.

5. A method for detecting the state of a tray of a tilting tray sorting machine according to claim 4, characterized in that: The calculation of the super-area type includes calculating a distance L, wherein the distance L represents a deviation distance between the package (12) and the pallet (10) on the upper side. When the distance L is greater than a set threshold, the state of the package (12) is judged to be IBB, and the IBB state indicates that the package (12) is stuck between two pallet carts.

6. A method for detecting the state of a tray of a tilting tray sorting machine according to claim 5, characterized in that: The calculation of the super-area type also includes calculating a distance K, wherein the distance K represents the deviation distance between the package (12) and the pallet (10) on the left side. When the distance K is greater than a set threshold, the state of the package (12) is judged to be IOB, and the IOB state indicates that the package (12) exceeds the edge of the pallet trolley.

7. The method for detecting the state of a tray of a tilting tray sorting machine according to claim 1, characterized in that: In step six, the calculation formula for calculating the intersection area of ​​the package (12) and the side plate (11) is: In the formula, S2 represents the ratio of the minimum circumscribed rectangle area of ​​the side panel, A represents the area occupied by the minimum circumscribed rectangle of the package, and C represents the area occupied by the minimum circumscribed rectangle of the side panel; when S2 is greater than the preset threshold, the package (12) is determined to be stuck in the side panel.

Images