Jump chain identification method and system based on visual inspection and computer equipment

Abstract

The invention relates to the technical field of machine vision detection, and discloses a skip chain identification method and system based on vision detection and computer equipment. The method comprises the following steps: connecting every two identified targets of at least two chain rings, and calculating the angle of the connecting lines to obtain a chain angle; if the chain angle is lower than the angle threshold value, the state is normal, and if the chain angle is higher than the angle threshold value, the state is chain jumping. In this way, it is possible to provide highly accurate determination and analysis when processing a large amount of data by means of a precise algorithm and model. Compared with manual judgment, higher consistency and stability can be kept, and the possibility of human errors is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of machine vision detection, and in particular to a chain skipping recognition method, system and computer equipment based on vision detection. Background Art

[0002] The cameras installed on the ship can record and monitor the situation near the anchor in real time. These cameras can capture the movement trajectory of the anchor chain and possible abnormal situations. The existing detection method mainly relies on manual viewing of the chain's operating status in the camera monitoring screen to make judgments, such as sudden loosening of the chain or abnormal position changes. Since manual judgment is easily affected by individual experience, emotions and subjective judgment, the consistency and accuracy of the judgment results are unstable. In addition, manual judgment requires the operator to observe the image, analyze the situation and make decisions, and the response speed is slow, which may cause delays or miss important information. Moreover, the persistence and patience of manual judgment may be affected by factors such as fatigue and distraction. It can be seen that the manual judgment method in the prior art has the disadvantages of unstable accuracy, slow response speed and high cost. Summary of the invention

[0003] The present invention provides a chain skipping identification method, system and computer equipment based on visual detection to solve the problems in the above-mentioned background technology.

[0004] In order to achieve the above object, the present invention is implemented by the following technical solutions:

[0005] In a first aspect, the present invention provides a chain skipping identification method based on visual detection, comprising:

[0006] S1: Obtain the chain running status video during the anchoring process and detect the chain lowering position in each video frame;

[0007] S2: Based on the chain lowering position, identify and output N specified feature points of the chain ring in each video frame;

[0008] S3: when it is recognized that there are at least two chain links in the video frame, calculating the angle difference based on N designated feature points of the at least two chain links;

[0009] S4: Determine the chain jumping state according to the angle difference.

[0010] Optionally, S2 includes:

[0011] Based on the chain lowering position, the feature point detection algorithm model is used to identify and output N specified feature points of the chain ring in each video frame and the coordinates of the minimum rectangular corner points of the chain ring;

[0012] Determine the range of the chain loop based on the coordinates of the minimum rectangular corner points of the chain loop;

[0013] It is determined whether the N specified feature points are within the range of the chain ring. If so, it is considered that the N specified feature points of the chain ring are successfully identified.

[0014] Optionally, in S2, based on the chain lowering position, a feature point detection algorithm model is used to identify and output N designated feature points of the chain ring in each video frame and the coordinates of the minimum rectangular corner points of the chain ring, including:

[0015] Each video frame is annotated based on the chain lowering position according to the set annotation method, and the coordinates of the minimum rectangular corner points of the chain ring and the N specified feature points are determined according to the annotation results.

[0016] Optionally, setting the marking method includes:

[0017] Identify and determine the center point of the chain loop in the video frame;

[0018] Mark the entire chain ring with a rectangular frame to obtain the coordinates of the minimum rectangular corner points of the chain ring;

[0019] The positions above the center point, the center point, and the position below the center point are marked once each to ensure that when the chain ring object is identified, there will be corresponding N specified feature points.

[0020] Optionally, S3 includes:

[0021] When it is recognized that there are at least two chain rings in the video frame, determining a target chain ring and a chain ring adjacent to the target chain ring;

[0022] Connect the N specified feature points of the target chain ring with the N specified feature points of the adjacent chain rings in a one-to-one correspondence;

[0023] The angle difference is determined based on the correspondingly connected N designated feature points.

[0024] Optionally, determining the angle difference based on the correspondingly connected N designated feature points includes:

[0025] Calculate the average angle of the N feature points currently connected;

[0026] The angle difference is obtained by comparing the average angle of the N feature points currently connected to each other with the average angle of the previous video frame.

[0027] Optionally, S4 includes:

[0028] If the angle difference is lower than the angle threshold, it is a normal state, and if the angle difference is higher than the angle threshold, it is a chain skipping state.

[0029] In a second aspect, the present application provides a chain skipping identification system based on visual detection, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method of the first aspect when executing the computer program.

[0030] In a third aspect, the present application provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method of the first aspect when executing the computer program.

[0031] Beneficial effects:

[0032] The chain skipping identification method based on visual detection provided by the present invention connects the targets of at least two chain links identified in pairs and calculates the angle of the connection to obtain the chain angle; if the chain angle is lower than the angle threshold, it is in a normal state, and if the chain angle is higher than the angle threshold, it is in a chain skipping state. In this way, it is possible to provide highly accurate judgment and analysis when processing large amounts of data through precise algorithms and models. Compared with manual judgment, it can maintain higher consistency and stability and reduce the possibility of human error.

[0033] In a further solution, the chain is identified through data annotation methods, which can improve the accuracy of recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 The figure is a flow chart of a chain skipping identification method based on visual detection according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0035] The technical solution of the present invention is described clearly and completely below. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the usual meanings understood by persons with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "one" do not indicate quantity restrictions, but indicate the existence of at least one. Words such as "connect" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship also changes accordingly.

[0037] It should be understood that the existing detection method is to manually check the running status of the chain in the camera monitoring screen to judge whether the chain suddenly loosens or changes in position abnormally. This is greatly affected by human factors, and the judgment accuracy is poor in the long run. Based on this, the present application provides a chain skipping identification method, system and computer device based on visual detection.

[0038] See also Figure 1 , the present application provides a chain skipping identification method based on visual detection, comprising:

[0039] S1: Obtain the chain running status video during the anchoring process and detect the chain lowering position in each video frame.

[0040] In this step, a video of the chain running state during the anchoring process is captured at a fixed position by any camera device, wherein the any camera device may be a camera, a sensor or an image acquisition device, which is only used as an example and is not limited thereto.

[0041] S2: Based on the chain lowering position, identify and output N specified feature points of the chain ring in each video frame.

[0042] In this step, the value of N can be 3 or 4. The specific value is the obvious feature points selected according to the actual imaging effect to facilitate algorithm recognition. The more the number, the more accurate it is. This is only an example and is not limited.

[0043] S3: When it is recognized that there are at least two chain links in the video frame, an angle difference is calculated based on N designated feature points of the at least two chain links.

[0044] S4: Determine the chain jumping state according to the angle difference.

[0045] The above-mentioned chain skipping identification method based on visual detection connects the targets of at least two chain links identified in pairs and calculates the angle of the connection to obtain the chain angle; if the chain angle is lower than the angle threshold, it is in a normal state, and if the chain angle is higher than the angle threshold, it is in a chain skipping state. In this way, precise algorithms and models can be used to provide highly accurate judgment and analysis when processing large amounts of data. Compared with manual judgment, it can maintain higher consistency and stability and reduce the possibility of human error.

[0046] Optionally, S2 includes:

[0047] Based on the chain lowering position, the feature point detection algorithm model is used to identify and output N specified feature points of the chain ring in each video frame and the coordinates of the minimum rectangular corner points of the chain ring;

[0048] Determine the range of the chain loop based on the coordinates of the minimum rectangular corner points of the chain loop;

[0049] It is determined whether the N specified feature points are within the range of the chain ring. If so, it is considered that the N specified feature points of the chain ring are successfully identified.

[0050] In one example, based on the chain lowering position, S2 uses a feature point detection algorithm model to identify and output N specified feature points of the chain ring in each video frame and the coordinates of the minimum rectangular corner points of the chain ring, including:

[0051] Each video frame is annotated based on the chain lowering position according to the set annotation method, and the coordinates of the minimum rectangular corner points of the chain ring and the N specified feature points are determined according to the annotation results.

[0052] Specifically, the marking method includes:

[0053] Identify and determine the center point of the chain loop in the video frame;

[0054] Mark the entire chain ring with a rectangular frame to obtain the coordinates of the minimum rectangular corner points of the chain ring;

[0055] The positions above the center point, the center point, and the position below the center point are marked once each to ensure that when the chain ring object is identified, there will be corresponding N specified feature points.

[0056] In this way, identifying chains through data labeling methods can improve the accuracy of identification.

[0057] Optionally, S3 includes:

[0058] When it is recognized that there are at least two chain rings in the video frame, determining a target chain ring and a chain ring adjacent to the target chain ring;

[0059] Connect the N specified feature points of the target chain ring with the N specified feature points of the adjacent chain rings in a one-to-one correspondence;

[0060] The angle difference is determined based on the correspondingly connected N designated feature points.

[0061] Specifically, determining the angle difference based on the correspondingly connected N designated feature points includes:

[0062] Calculate the average angle of the N feature points currently connected;

[0063] The angle difference is obtained by comparing the average angle of the N feature points currently connected to each other with the average angle of the previous video frame.

[0064] In this implementation, the angle difference is obtained by using the angle average value as a basis for judgment, which can improve the stability of angle acquisition and reduce errors and occasional angle deviations.

[0065] Optionally, S4 includes:

[0066] If the angle difference is lower than the angle threshold, it is a normal state, and if the angle difference is higher than the angle threshold, it is a chain skipping state.

[0067] In specific implementation, the angle threshold is based on actual experience and equipment, and presents data differences according to different scenarios.

[0068] The present application also provides a chain skipping identification system based on visual detection, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the steps of the above method are implemented when the processor executes the computer program.

[0069] The chain skipping identification system based on visual detection can implement various embodiments of the chain skipping identification method based on visual detection described above, and can achieve the same beneficial effects, which will not be described in detail here.

[0070] The present application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0071] The computer device can implement various embodiments of the above-mentioned chain skipping identification method based on visual detection, and can achieve the same beneficial effects, which will not be described in detail here.

[0072] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A chain skipping identification method based on visual detection, characterized in that: include: S1: Obtain the chain running status video during the anchoring process and detect the chain lowering position in each video frame; S2: Based on the chain lowering position, identify and output N designated feature points of the chain loop in each video frame; S3: when it is recognized that there are at least two chain loops in the video frame, calculating the angle difference based on N designated feature points of the at least two chain loops; S4: Determine the chain jumping state according to the angle difference.

2. The chain skipping identification method based on visual detection according to claim 1, characterized in that: The S2 includes: Based on the chain lowering position, identify and output N designated feature points of the chain loop in each video frame and the coordinates of the minimum rectangular corner points of the chain loop through a feature point detection algorithm model; Determining the range of the chain loop based on the coordinates of the minimum rectangular corner points of the chain loop; It is determined whether the N designated feature points are within the range of the chain ring. If so, it is considered that the N designated feature points of the chain ring are successfully identified.

3. The chain skipping identification method based on visual detection according to claim 2, characterized in that: In S2, based on the chain lowering position, the feature point detection algorithm model is used to identify and output N designated feature points of the chain ring in each video frame and the coordinates of the minimum rectangular corner points of the chain ring, including: Each video frame is annotated based on the chain lowering position according to a set annotation method, and N designated feature points and the coordinates of the minimum rectangular corner points of the chain ring are determined according to the annotation results.

4. The chain skipping identification method based on visual detection according to claim 3 is characterized in that: The setting marking method includes: Identify and determine the center point of the chain loop in the video frame; Mark the entire chain ring with a rectangular frame to obtain the coordinates of the minimum rectangular corner points of the chain ring; The position above the center point, the position of the center point, and the position below the center point are each marked once, to ensure that there are corresponding N designated feature points when the chain ring object is identified.

5. The method for identifying chain skipping based on visual detection according to claim 1, characterized in that: The S3 includes: When it is recognized that there are at least two chain rings in the video frame, determining a target chain ring and a chain ring adjacent to the target chain ring; Connect the N specified feature points of the target chain ring with the N specified feature points of the adjacent chain rings in a one-to-one correspondence; The angle difference is determined based on the correspondingly connected N designated feature points.

6. The method for identifying chain skipping based on visual detection according to claim 5, characterized in that: Determining the angle difference based on the correspondingly connected N designated feature points includes: Calculate the average angle of the N feature points currently connected; The angle difference is obtained by comparing the average angle of the N feature points currently connected to each other with the average angle of the previous video frame.

7. The method for identifying chain skipping based on visual detection according to claim 1, characterized in that: The S4 includes: If the angle difference is lower than the angle threshold, it is a normal state; if the angle difference is higher than the angle threshold, it is a chain skipping state.

8. A chain skipping identification system based on visual detection, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method described in any one of claims 1 to 7 are implemented.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

Images