A conveying belt tearing visual monitoring method based on double-camera cooperation

Abstract

Traditional conveyor belt tear detection methods suffer from drawbacks such as complex installation, susceptibility to conveyor belt wear, and low detection sensitivity. Non-contact detection methods, particularly single-camera detection, have limited fields of view and are susceptible to interference, leading to missed or false detections. Dual-camera solutions often use cameras with similar functions for image acquisition and rely on software algorithms for synchronous data recognition, lacking a differentiated division of labor. Therefore, this invention proposes a visual detection and relay control device and method for conveyor belt tears based on dual-camera collaboration. Two industrial cameras are used for conveyor belt tear detection and triggering, as well as tear image capture, tear size identification, and early warning. This achieves a balance between high-speed early warning and high-precision imaging, resolving the contradiction of insufficient frame rate and resolution in single-camera systems. Laser structured light detection and direct image recognition complement each other, expanding the tear detection range and improving the comprehensive detection capabilities for early, minute, surface, and deep tears. Relay hardware triggering provides strong real-time performance and excellent anti-interference capabilities.

Description

Technical Field

[0001] This invention relates to the field of conveyor belt safety inspection technology, specifically to a visual inspection and relay control device and method for conveyor belt tear detection based on dual-camera collaboration, applicable to real-time tear detection of conveyor belts. Background Technology

[0002] Traditional conveyor belt tear detection mainly falls into two categories: contact detection and non-contact detection. Contact detection suffers from drawbacks such as complex installation, susceptibility to conveyor belt wear, and low detection sensitivity, making it unsuitable for high-speed conveyor belt environments. In non-contact detection, existing machine vision solutions with single cameras have limited fields of view and are prone to missed or false detections due to interference from dust, light spots, and debris on the conveyor belt surface. Dual-camera solutions often use cameras with similar functions for image acquisition, relying on software algorithms for synchronous data recognition, without establishing a differentiated division of labor mechanism. Therefore, this invention proposes a conveyor belt tear visual detection and relay control device and method based on dual-camera collaboration. Two cameras respectively detect and trigger conveyor belt tears, capture tear images, identify tear size, and provide early warning. This achieves a balance between high-speed early warning and high-precision imaging, resolving the contradiction that single-camera frame rate and resolution cannot be simultaneously met. Laser structured light detection and direct image recognition complement each other, expanding the tear detection range and improving the comprehensive detection capability for early, small, surface, and deep tears. Relay hardware triggering collaboration provides strong real-time performance and outstanding anti-interference capabilities. Summary of the Invention

[0003] The present invention aims to provide a visual detection system for conveyor belt tearing based on dual-camera collaboration, which can achieve accurate detection, rapid triggering, judgment and identification and triggering early warning of conveyor belt tearing, thereby improving the safety and reliability of conveyor belt operation.

[0004] According to one embodiment of the present invention, the device includes a conveyor belt body, a dual-camera acquisition unit, a host computer, a USB relay, a linear laser, and a bracket, and is divided into a tear detection module, a camera triggering module, an image processing module, a size recognition module, and a threshold warning module.

[0005] The dual-camera acquisition unit includes a first industrial camera and a second industrial camera. The first industrial camera is fixed above the working surface of the conveyor belt, with its lens tilted downwards towards the surface of the conveyor belt. The second industrial camera is arranged on the downstream side of the conveyor belt's running direction, with its lens in an oblique, downward-looking posture. Both are area array cameras.

[0006] The linear laser is installed upstream of the first industrial camera, and the laser is installed perpendicular to the direction of the conveyor belt's movement.

[0007] The aluminum profile is fixed to the upper surface of the bracket by C-clamps at both ends. The two threaded holes on the lower surface of the first industrial camera and the central groove of the aluminum profile are fixed with bolts. The threaded part of the bolt is screwed into the threaded hole of the camera, and the head is fixed inside the central groove of the aluminum profile. The two sides of the camera are fixed to the aluminum profile by angle brackets, bolts and T-nuts.

[0008] The laser is fixed above the aluminum profile by angle brackets, bolts and T-nuts. The aluminum profile is fixed to the upper surface of the bracket by C-clamps at both ends. The laser line is perpendicular to the direction of the conveyor belt.

[0009] The bracket is used to fix the first industrial camera and the line laser in a suitable position on the conveyor belt.

[0010] The second industrial camera is fixed to the aluminum profile center groove via two threaded holes on its lower surface. The threaded shank of the bolt is screwed into the threaded hole of the camera, and the head is fixed inside the center groove of the aluminum profile. The two sides of the camera are fixed to the aluminum profile via angle brackets, bolts, and T-nuts. The aluminum profile is fixed to the angle steel above the conveyor via C-clamps.

[0011] The host computer establishes a communication connection with the USB relay module through its own USB interface, the power module is connected to the input terminal of the USB relay, and the output terminal of the USB relay is connected to the trigger terminal of the second industrial camera.

[0012] The USB relay input is powered by an external power module, and the output is connected to the camera's trigger port via a terminal block.

[0013] The power supply unit provides power to the industrial camera, which is connected to the switch via a network cable. At the same time, the switch is connected to the network port of the host computer via a network cable.

[0014] The tear detection module is used to detect whether the laser line is interrupted. The first industrial camera monitors the gaps in the entire laser line on the surface of the conveyor belt in real time. Once a gap is detected in the laser line, it is identified as a tear in the conveyor belt.

[0015] The camera trigger module is used to trigger the second industrial camera. After the first industrial camera detects that the conveyor belt has torn, it triggers the relay to switch between high and low levels. At the same time, the relay triggers the second industrial camera to capture the torn image.

[0016] The image processing module is used to process the conveyor belt images transmitted to the host computer. The host computer preprocesses the images, extracts tear features from the images using a set tear detection algorithm, outputs the tear judgment results from the second industrial camera, selects the area where the tear occurs, and outputs the confidence level.

[0017] The size recognition module is used to identify the length and width of the tear. By using camera calibration parameters, the length and width pixel dimensions of the conveyor belt tear image are converted into physical dimensions.

[0018] The threshold warning module is used to pre-set the length and width thresholds of tears in the detection system. Tears below the threshold will only output the pixel and physical dimensions of the tear length and width. Tears exceeding the threshold will output a pop-up warning in addition to tear information.

[0019] Traditional conveyor belt tear detection methods suffer from drawbacks such as complex installation, susceptibility to conveyor belt wear, and low detection sensitivity. Non-contact detection methods, particularly single-camera detection, have limited fields of view and are susceptible to interference, leading to missed or false detections. Dual-camera solutions often use cameras with similar functions for image acquisition and rely on software algorithms for synchronous data recognition, lacking a differentiated division of labor. Therefore, this invention proposes a visual detection and relay control device and method for conveyor belt tears based on dual-camera collaboration. Two industrial cameras are used for conveyor belt tear detection and triggering, as well as tear image capture, tear size identification, and early warning. This achieves a balance between high-speed early warning and high-precision imaging, resolving the contradiction of insufficient frame rate and resolution in single-camera systems. Laser structured light detection and direct image recognition complement each other, expanding the tear detection range and improving the comprehensive detection capabilities for early, minute, surface, and deep tears. Relay hardware triggering provides strong real-time performance and excellent anti-interference capabilities. Attached Figure Description

[0020] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings required in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0021] Figure 1 This is a field layout diagram of the dual-camera structure described in this invention.

[0022] Figure 2 This is a schematic diagram of the structure of the conveyor belt tear visual monitoring system based on dual cameras according to the present invention.

[0023] Figure 3 This is an automatic detection flowchart of the dual-camera-based visual monitoring system for conveyor belt tearing according to the present invention.

[0024] Figure 4 This is a schematic diagram illustrating the control of the USB relay, host computer, and industrial camera described in this invention.

[0025] In the attached diagram: 1. Bracket; 2. Aluminum profile; 3. C-clamp; 4. Laser; 5. First industrial camera; 6. Angle bracket; 7. Bolt; 8. T-nut; 9. Second industrial camera. Detailed Implementation

[0026] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] This monitoring system includes a conveyor belt body, a dual-camera acquisition unit, a host computer, and a USB relay. It is divided into a tear detection module, a camera triggering module, an image processing module, and a threshold warning module. The tear detection module detects whether the laser line is interrupted; the camera triggering module triggers the second industrial camera 9 via the relay; the image processing module processes the conveyor belt images transmitted to the host computer; the size recognition module identifies the length and width of the tear; and the threshold warning module pre-sets tear length and width thresholds in the detection system.

[0028] The host computer establishes a communication connection with the USB relay module via its own USB interface to receive the preview image from the first industrial camera 5 during the inspection process. Simultaneously, the USB relay module sends control commands to control the second industrial camera 9 to capture the torn image.

[0029] The area scan industrial camera transmits images to the industrial control computer in real time via network port.

[0030] In image processing and tear recognition, the host computer preprocesses the conveyor belt image by removing dust and noise interference through Gaussian filtering; it then calls the tear detection algorithm model to detect tears in the preprocessed image and outputs the length, width, and confidence level of the suspected tear area. At the same time, if the tear exceeds the set threshold size, a tear pop-up warning is output.

[0031] Figure 1The field layout is a dual-camera setup. The first industrial camera 5 is positioned diagonally above the working surface of the conveyor belt, tilting downwards to capture images of the conveyor belt surface, used to acquire laser line images of the entire conveyor belt area. The second industrial camera 9 is positioned downstream of the conveyor belt in the direction of travel, tilting downwards to capture tear images. The two cameras are spaced apart along the longitudinal direction of the conveyor belt, jointly covering the critical areas of the conveyor belt prone to tearing. The first industrial camera 5 is fixed on the bracket 1, the laser 4 is vertically installed in the direction of travel of the conveyor belt, located upstream of the first industrial camera 5, and the second industrial camera 9 is fixed together with the angle steel above the frame, located downstream of the first industrial camera 5.

[0032] Figure 2 This is a schematic diagram of the structure of a visual monitoring system, including a tear detection module, a camera triggering module, an image processing module, and a threshold warning module.

[0033] Figure 3 This is a flowchart of the automatic conveyor belt tear detection process. The process includes camera image acquisition, a USB relay activation command from the host computer, relay activation, image preprocessing and tear feature recognition, tear result determination, tear confidence threshold determination, fault confirmation, and system reset, completing one detection and control loop. The confidence threshold can be flexibly adjusted according to industrial site conditions to ensure the reliability of the detection results. If the confidence threshold determination fails, the process returns to the camera image acquisition step to continue real-time detection.

[0034] Figure 4 This diagram illustrates the control of a USB relay, a host computer, and industrial cameras. The relay connects to the host computer's USB port and the trigger terminal of the second industrial camera 9. A new power module, consisting of a power supply and a step-down module, provides external power to the relay. The two industrial cameras are connected to the power supply and, via a switch, to the host computer. They are responsible for transmitting image information and triggering the tearing image capture.

[0035] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications should also be considered to fall within the scope of protection of this application.

Claims

1. A visual monitoring method for conveyor belt tearing based on dual-camera collaboration, characterized in that, The dual-camera acquisition unit includes a first industrial camera (5) and a second industrial camera (9), both of which are area array cameras, and a line laser (4). The first industrial camera (5) is fixed above the working surface of the conveyor belt, with its lens tilted downwards towards the surface of the conveyor belt; the second industrial camera (9) is arranged downstream of the conveyor belt in the direction of travel, with its lens tilted downwards. The line laser (4) is installed upstream of the first industrial camera (5), and the laser (4) is installed vertically in the direction of travel of the conveyor belt.

2. The detection method according to claim 1, characterized in that, The tear detection module uses a first industrial camera (5) to detect the interruption of the laser line on the conveyor belt surface in real time. The camera trigger module triggers a second industrial camera (9) via a relay when the first industrial camera (5) detects an interruption in the laser line. The image processing module processes the conveyor belt image transmitted to the host computer, selects the area where the tear occurred, and outputs the confidence level of the detected tear image. The size recognition module identifies the pixel dimensions and physical dimensions of the tear length and width. The threshold warning module provides a pop-up warning for tear images exceeding the set tear length and width thresholds.

3. The detection method according to claim 1, characterized in that, The camera trigger module includes a USB relay that triggers the second industrial camera (9) to capture a torn image. When the first industrial camera (5) detects a gap in the laser line on the surface of the conveyor belt, it triggers the relay to generate a high-low level transition. The relay also serves as an external trigger for the second industrial camera (9), triggering the second industrial camera (9) to capture a torn image.

4. The detection method according to claim 1, characterized in that, The method specifically includes the following steps: Step 1: The first industrial camera (5) acquires laser line images of the surface of the running conveyor belt in real time; Step 2: Transmit the identified laser line image information to the host computer via network cable; Step 3: The host computer judges the laser line information. If the laser line is identified as an interrupt, the relay is triggered; otherwise, the first industrial camera (5) continues to perform the judgment and recognition work. Step 4: After the relay is triggered, the second industrial camera (9) is also triggered to capture images. Step 5: The captured image is transmitted to the host computer for tear image recognition and tear size output.

Images