A machine vision-based safety distance detection method for a power plant ship unloader cab

Abstract

The application provides a kind of based on machine vision's power plant ship unloader cab safety distance detection method, including S1: collection visible light image and infrared image;S2: the detection work of figure to visible light image is carried out, and check straight line segment;S3: determine the number of obtained figure is greater than or equal to three, calculate figure center distance x, according to the focal length of camera conversion distance X between figure, and determine the number of obtained straight line is greater than or equal to three, calculate straight line length d, according to the focal length of camera conversion straight line length D;S4: determine the target value of obtained X is greater than or equal to three, and determine the target value of obtained D is greater than or equal to three, merge processing data, carry out the determination of safety distance;Its technical key points are: by using dual-spectrum camera respectively collection visible light image and infrared image obtains temperature data, by data processor all-in-one machine processing, can realize the real-time detection of the distance from ship unloader cab to coal ship, and timely prompt driver to pay attention to safety.

Description

Technical Field

[0001] This invention belongs to the field of ship unloader safety technology, specifically a machine vision-based method for detecting the safe distance from the cab of a ship unloader in a thermal power plant. Background Technology

[0002] Bridge unloaders, as large-scale unloading equipment, play a vital role in coastal thermal power plants. As an important unloading tool, the performance of the bridge unloader directly affects the transportation of coal; therefore, the safety and operational stability of the unloader are of paramount importance. The unloader mainly consists of a material-grabbing mechanism, a traveling mechanism, and a cab operating mechanism. The cab is suspended on the main beam. Utilizing a continuous conveyor system, the unloader head, which lifts bulk materials, either with self-loading capabilities or equipped with material-grabbing and feeding devices, continuously lifts bulk materials from the ship's hold, unloads them onto the boom or frame, and transports them to the main conveyor on shore. This significantly improves unloading efficiency, reduces dust pollution, maintains environmental cleanliness, and is highly efficient and environmentally friendly.

[0003] Existing bridge-type ship unloaders unload coal by working at height. During operation, the operator is located in the cab, and their effective field of vision is only 120°. The operator has a large blind spot. At the same time, when operating in low visibility environments such as rain, fog, or at night, the obstacles to the driver's operation are even greater, which can easily lead to safety accidents. Summary of the Invention

[0004] To overcome existing shortcomings, this application provides a machine vision-based method for detecting the safe distance from the cab of a ship unloader in a thermal power plant. By using a dual-spectrum camera to acquire visible light and infrared images to obtain temperature data, which is then processed by a data processor, the method can achieve real-time detection of the distance between the ship unloader cab and the coal ship. When the distance between the cab and the coal ship is less than the safe distance, the driver is alerted to pay attention to safety, ensuring the safe operation of the ship unloader. This method solves the problem that existing bridge ship unloaders operate at height, resulting in a large blind spot for the operator in the cab, which is also greatly affected by low visibility environments such as rain, fog, and night, making it easy for safety accidents to occur during the unloading process.

[0005] The technical solution adopted by the embodiments of this application to solve its technical problem is: A machine vision-based method for detecting safe distances from the operator's cab of a ship unloader in a thermal power plant, including... Step 1: Simultaneously acquire visible light and infrared images, preprocess the obtained visible light image using Gaussian filtering, and retain the infrared image data; Step 2: Detect patterns in the visible light image, use different colors to filter the number of valid patterns, retain thermal imaging temperature data, and check straight line segments; Step 3: Determine if the number of obtained graphics is greater than or equal to three. If the result is less than three, return to the visible light image acquisition stage. If the result is greater than or equal to three, calculate the center distances x1, x2, x3, x4, and x5 of the graphics, and convert the distances between the graphics to X1, X2, X3, X4, and X5 based on the camera's focal length. Also, determine if the number of obtained straight lines is greater than or equal to three. If the result is less than three, return to the infrared image acquisition stage. If the result is greater than or equal to three, calculate the lengths d1, d2, d3, and d4 of the straight lines, and convert the lengths D1, D2, D3, and D4 based on the camera's focal length. Step 4: Determine whether the number of X values ​​obtained is greater than or equal to three, and determine whether the number of D values ​​obtained is greater than or equal to three. When the number of X and D values ​​is greater than or equal to three, merge the data. Step 5: Average the obtained data to determine the distance from the center position to the driver, and then determine the safe distance.

[0006] In one possible implementation, the graphics processed in steps two and three include both ellipses and circles.

[0007] In one possible implementation, the thermal imaging temperature data retained in step two is between 45 and 55 degrees Celsius.

[0008] In one possible implementation, the colors used to filter the number of valid graphics in step two include white, red, blue, and green.

[0009] In one possible implementation, when the number of X is less than three, the camera lens tracks the target and magnifies the image, returning to the visible light image acquisition stage; when the number of D is less than three, it directly returns to the infrared image acquisition stage.

[0010] In one possible implementation, the target value of X in step four is 100 mm and 140 mm, and the target value of D is 100 mm.

[0011] In one possible implementation, in step five, if the distance from the center position to the driver is less than the safe distance, the driver's cab is in danger; if the distance from the center position to the driver is greater than the safe distance, the driver's cab is safe. Regardless of the result, the process returns to the visible light image acquisition and infrared image acquisition stage.

[0012] In one possible implementation, step one includes a light-emitting and heat-generating element made of non-metallic material, which is placed in a regular quadrilateral shape at a high position on the coal ship facing the unloader's cab to assist in infrared image acquisition.

[0013] In one possible implementation, the temperature of the light-emitting and heat-generating element is controlled between 43 and 45 degrees Celsius, and can reach 55 degrees Celsius when the light-emitting and heat-generating element is exposed to sunlight.

[0014] In one possible implementation, step one includes a dual-spectrum camera, which is installed inside the driver's cab. The output of the dual-spectrum camera is connected to a data processor integrated machine. The model of the dual-spectrum camera is Hikvision DS-2TD55XXT-XX / W.

[0015] The beneficial effects of this application are as follows: First, in this solution, temperature data is obtained by using a dual-spectrum camera to collect visible light and infrared images respectively. The data is then processed by an integrated data processor, which enables real-time detection of the distance between the unloader's cab and the coal ship. When the distance between the cab and the coal ship is less than the safe distance, the driver is alerted to pay attention to safety, thus ensuring the safe operation of the unloader. Secondly, this solution uses the principle of pinhole imaging to process the visible light image of the target element and the temperature data of the thermal imager with machine vision technology, identifies the distance from the target element to the camera, and determines the distance from the coal carrier to the unloader's cab, thus providing a foundation for the safety of the cab. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the detection process of the present invention; Figure 2 This is a schematic diagram illustrating the pinhole imaging principle of the present invention. Detailed Implementation

[0017] The technical solution in this application embodiment is to solve the problems mentioned in the background art, and the overall idea is as follows: Example 1: This embodiment introduces a machine vision-based method for detecting safe distances from the cab of a ship unloader in a thermal power plant. For details, please refer to... Figures 1-2 As shown, it includes: S1: Simultaneously acquire visible light and infrared images, preprocess the obtained visible light image using Gaussian filtering, and retain the infrared image data; The target component is placed on the ship unloader, and a dual-spectrum camera captures images of it. A visible light video image is acquired every second, and the video stream is decoded into image information and saved in matrix format. Simultaneously, full-screen temperature data from a thermal imager is acquired every second and saved in matrix format.

[0018] At the same time, both visible light and infrared images are acquired at a frequency of 1 frame per second; S2: Perform graphic detection on the visible light image, use different colors to filter the number of valid graphics, retain thermal imaging temperature data, and check straight line segments; The detected shapes include ellipses and circles, and are filtered using four colors: white, red, blue, and green, to obtain the number of valid ellipses or circles, while retaining the thermal imaging temperature data from 45° to 55°. S3: Determine if the number of obtained graphics is greater than or equal to three. If the result is less than three, return to the visible light image acquisition stage. If the result is greater than or equal to three, calculate the center distances x1, x2, x3, x4, and x5 of the graphics, and convert the distances between the graphics to X1, X2, X3, X4, and X5 based on the camera's focal length. Also, determine if the number of obtained straight lines is greater than or equal to three. If the result is less than three, return to the infrared image acquisition stage. If the result is greater than or equal to three, calculate the lengths d1, d2, d3, and d4 of the straight lines, and convert the lengths D1, D2, D3, and D4 based on the camera's focal length. The actual target element has four circles. When more than three ellipses or circles are detected, it indicates that a suspected target element has been found. In the straight line segment inspection, due to the arrangement of the light-emitting and heat-generating elements in a square shape, when more than three straight lines are detected, it indicates that a suspected target element has been found. S4: Determine whether the obtained target value X is greater than or equal to three, and determine whether the obtained target value D is greater than or equal to three. When the number of X and D is greater than or equal to three, merge the data. Among them, the center distance between suspected target ellipses or circles is calculated based on the focal length. When the center distance is 1.0m or 1.44m, it is confirmed to be a target element. The suspected target line is calculated based on the focal length. When the line is 1.0m, it is confirmed to be a target element. S5: The average data is used to determine the distance from the center position to the driver, and a safe distance is determined. The camera used in the cab is a dual-spectrum camera, and its output is connected to a data processor. When the camera collects visible light images and thermal imaging temperature data, and combines the data from the visible light and thermal imager, a danger warning is issued when the distance between the target element and the camera, i.e. the cab, is less than the set value. Meanwhile, the model of the dual-spectrum camera can be Hikvision's DS-2TD55XXT-XX / W.

[0019] By adopting the above technical solution: The above design uses a dual-spectrum camera to acquire visible light and infrared images to obtain temperature data. It then uses four different colors to filter the number of valid ellipses and circles and detects the number of straight line segments. After processing by an integrated data processor, it performs a safe distance determination. This allows for real-time detection of the distance between the unloader's cab and the coal ship. When the distance between the cab and the coal ship is less than the safe distance, the driver is alerted to pay attention to safety, ensuring the safe operation of the unloader.

[0020] Example 2: Based on Example 1, this example introduces the basic principle of a machine vision-based method for detecting safe distances in the cab of a ship unloader in a thermal power plant, including object distance u, image distance v, object height D, and image height d. In calculating the focal length f, let 1 / u + 1 / v = 1 / f, then we get D = f / (1 - f / d). We can then calculate the actual size of the object based on the size of the image and the focal length, which is simple and convenient.

[0021] By adopting the above technical solution: The above design, by setting 1 / u+1 / v=1 / f, yields D=f / (1-f / d), which calculates the actual size of the object based on the image size and focal length. This helps reduce the processing difficulty of the integrated data processor, improves processing efficiency, and ensures the safety of the driver during coal unloading operations in the cab.

[0022] Finally, it should be noted that the above embodiments are merely examples for clearly illustrating the present invention and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A machine vision-based method for detecting safe distances from the cab of a ship unloader in a thermal power plant, characterized in that, include: Step 1: Simultaneously acquire visible light and infrared images. Preprocess the visible light image using Gaussian filtering, while retaining the infrared image data. Step 2: Perform pattern detection on the visible light image. Use different colors to filter the number of valid patterns, retain thermal imaging temperature data, and check straight line segments. Step 3: Determine if the number of obtained patterns is greater than or equal to three. If the result is less than three, return to the visible light image acquisition stage. If the result is greater than or equal to three, calculate the center distances x1, x2, x3, x4, and x5 of the patterns. Convert the distances between patterns X1, X2, X3, X4, and X5 according to the camera's focal length. Also determine if the number of obtained straight lines is greater than or equal to three. If the result is less than three, return to the infrared image acquisition stage. If the result is greater than or equal to three, calculate the straight line length d1. d2, d3, d4, calculate the straight line lengths D1, D2, D3, D4 based on the camera focal length; Step 4: Determine whether the obtained target value X is greater than or equal to three, and determine whether the obtained target value D is greater than or equal to three. When the number of X and D is greater than or equal to three, merge the data; Step 5: Average the obtained data to obtain the distance from the center position to the driver, and determine the safe distance; The graphics processed in Step 2 and Step 3 include ellipses and circles; Specifically, a target element is set up. This target is a light-emitting and heat-generating element, which is made of non-metallic material and is placed in the form of a regular quadrilateral at a high position on the coal ship facing the unloader's cab to assist in infrared image acquisition. The actual target element has 4 circles. When more than 3 ellipses or circles are detected, it indicates that a suspected target element has been found.

2. The method for detecting safe distances from the cab of a ship unloader in a thermal power plant based on machine vision as described in claim 1, characterized in that: The thermal imaging temperature data retained in step two is between 45 and 55 degrees Celsius.

3. The method for detecting safe distances to the cab of a ship unloader in a thermal power plant based on machine vision as described in claim 1, characterized in that: The colors used to filter the number of valid images in step two include white, red, blue, and green.

4. The method for detecting safe distances to the cab of a ship unloader in a thermal power plant based on machine vision as described in claim 1, characterized in that: In step four, when the number of X is less than three, the camera lens tracks the target and magnifies the image, returning to the visible light image acquisition stage. When the number of D is less than three, it directly returns to the infrared image acquisition stage.

5. The method for detecting safe distances from the cab of a ship unloader in a thermal power plant based on machine vision as described in claim 1, characterized in that: In step four, the target values ​​for X are 100 mm and 140 mm, while the target value for D is 100 mm.

6. The method for detecting safe distances to the cab of a ship unloader in a thermal power plant based on machine vision as described in claim 1, characterized in that: In step five, if the distance from the center position to the driver is less than the safe distance, the driver's cab is in danger; if the distance from the center position to the driver is greater than the safe distance, the driver's cab is safe. Regardless of the result, the process returns to the visible light image acquisition and infrared image acquisition stage.

7. The method for detecting safe distances to the cab of a ship unloader in a thermal power plant based on machine vision as described in claim 1, characterized in that: The temperature of the light-emitting and heat-generating element is controlled between 43 and 45 degrees Celsius, and can reach 55 degrees Celsius when the light-emitting and heat-generating element is exposed to sunlight.

8. The method for detecting safe distances to the cab of a ship unloader in a thermal power plant based on machine vision as described in claim 1, characterized in that: Step one includes a dual-spectrum camera, which is installed inside the driver's cab, and the output of the dual-spectrum camera is connected to a data processor integrated machine.

Images