Detection method for cutting width of combine harvester

Abstract

The invention relates to a method for detecting the cutting width of a combine harvester. The method comprises the following steps that S1, a laser arranged on the combine harvester emits laser to circularly scan crops in front of a header; s2, continuously acquiring images of an area in front of the header through an industrial camera arranged on the combine harvester; s3, processing the image acquired by the industrial camera, and acquiring two straight lines corresponding to the cut edge; s4, according to the obtained two straight lines corresponding to the cut edge, the pixel distance between the two straight lines is solved, and the pixel distance is the pixel cut width; s5, the focal length and the object distance of the industrial camera are obtained, and the actual cutting width is calculated according to the focal length and the object distance, and S6, the vehicle speed of the combine harvester is obtained through a Hall sensor arranged on the combine harvester, and the cutting area of the combine harvester is calculated according to the obtained actual cutting width and the vehicle speed. According to the invention, the cutting area of the combine harvester can be rapidly obtained.

Description

Technical Field

[0001] The invention relates to the technical field of agricultural machinery, and in particular to a method for detecting the cutting width of a combine harvester. Background Art

[0002] Combine harvesters are of indispensable importance in modern agriculture. They not only improve agricultural production efficiency and harvesting quality, reduce labor intensity and resource consumption, but also promote the modernization of agriculture, increase farmers' income, and promote the sustainable development of agriculture. Determining the cutting area of ​​the combine harvester is crucial in agricultural production. However, since the combine harvester undertakes multiple tasks during operation, a large amount of debris will be generated at the operation site. At this time, the driver only uses human vision to judge the harvesting range in this harsh environment, which will cause a large error and it is difficult to obtain the accurate grain boundary. Therefore, real-time detection of the harvester's cutting width is one of the important tasks to realize the intelligent operation of the harvester and reduce the workload of workers.

[0003] In recent years, with the continuous development of smart agriculture, domestic and foreign researchers have conducted many studies on cutting width measurement. Among them, the most important research direction is non-contact measurement. The main feature of this method is the use of two ultrasonic sensors placed on both sides of the combine harvester header. When the combine harvester is operating, the sensor will emit ultrasonic signals and reflect them back after encountering the grain closest to the sensor. The sensor then receives these reflected signals and determines the distance by calculating the time difference between the transmitted and received signals. However, due to the thin stems of crops during actual planting, the signal may pass through or fail to return, resulting in measurement errors. Therefore, using a more accurate method to detect the cutting width of the harvester in real time is one of the important tasks to realize the intelligent operation of the harvester and reduce the workload of workers. Summary of the invention

[0004] In view of the defects of the prior art, an object of the present invention is to provide a method for detecting the cutting width of a combine harvester.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is: a method for detecting the cutting width of a combine harvester, comprising the following steps: Step S1: a laser device installed on the combine harvester emits laser light to perform a cyclic scan on the crops in front of the harvesting platform; Step S2: continuously acquiring images of the area in front of the harvesting platform through an industrial camera installed on the combine harvester; Step S3: processing the images acquired by the industrial camera to acquire two straight lines corresponding to the edges of the cutting swath; Step S4: calculating the pixel distance between the two straight lines based on the acquired two straight lines corresponding to the edges of the cutting swath, and the pixel distance is the pixel cutting swath width d1; Step S5: Obtain the focal length f and object distance D of the industrial camera. The object distance D is the distance between the installation position of the industrial camera and the center of the area scanned by the laser. Define the actual cutting width as d2. According to the principle of similar triangles: The actual cutting width is Step S6: The speed of the combine harvester is obtained by a Hall sensor arranged on the combine harvester, and the cutting area of ​​the combine harvester is calculated according to the obtained actual cutting width d2 and the speed.

[0006] Furthermore, the area cyclically scanned by the laser is a fan-shaped area, and the laser emitted by the laser is blue light.

[0007] Furthermore, in step S3, processing the image acquired by the industrial camera includes the following steps: denoising the image by means of median filtering, then performing color threshold segmentation on the image by using the difference between the blue light emitted by the laser transmitter and the crops and the ground, and then performing feature extraction on the extracted laser irradiation area, that is, performing edge detection, and processing the image by means of the Canny edge detection algorithm to highlight the edge portion of the actual cut in the laser irradiation area of ​​the image, using morphological processing to appropriately thicken overly thin edge lines, and then performing straight line detection, using the Hough transform straight line detection method to detect two straight lines corresponding to the cut edges from the edge image.

[0008] Furthermore, the two straight lines are the first straight line and the second straight line. The polar coordinate parameters of the first straight line and the second straight line in the image are detected by the Hough transform straight line detection method as (ρ1, θ1) and (ρ2, θ2), respectively. The polar coordinate equations of the first straight line and the second straight line are converted into rectangular coordinate equations. The polar coordinate equation of the first straight line is ρ1 = x1 cosθ1 + y1sinθ1, then the rectangular coordinate equation of the first straight line is The polar coordinate equation of the second straight line is ρ2 = x2cosθ2 + y2 sinθ2, so the rectangular coordinate equation of the second straight line is According to the rectangular coordinate equation of the first straight line, select the intercept point P1 (x1, y1) of the first straight line in the rectangular coordinate system, and according to the rectangular coordinate equation of the second straight line, select the intercept point P2 (x2, y2) of the second straight line in the rectangular coordinate system. Use the distance formula between the two points To calculate the distance between the two points, we can get the pixel distance between the two lines.

[0009] Furthermore, the intercept point P1(x1,y1) is the horizontal intercept point, and P2(x2,y2) is the horizontal intercept point.

[0010] Furthermore, the intercept point P1(x1,y1) is the vertical intercept point, and P2(x2,y2) is the vertical intercept point.

[0011] Furthermore, obtaining the object distance D of the industrial camera in step S5 includes the following steps: obtaining the height H1 of the installation position of the industrial camera from the horizontal ground and the shortest distance H2 between the vertical plane where the industrial camera is located and the vertical plane where the laser emitter hits the crop, and using the Pythagorean theorem of triangles to obtain the object distance

[0012] In step S6, calculating the cutting swath area of ​​the combine harvester includes the following steps: According to the acquired vehicle speed, the moving distance of the combine harvester within a certain period of time can be obtained, and the cutting swath area can be obtained by multiplying the moving distance by the actual cutting swath width d2.

[0013] Beneficial effects: The present invention adopts non-contact method, laser measurement technology, image recognition processing technology, etc., so as to realize real-time cutting width measurement of the harvester, and obtains the speed of the combine harvester in combination with the Hall sensor, so as to obtain the cutting area of ​​the combine harvester, reduce the burden on the driver, improve crop harvesting efficiency, provide relevant basis for my country's agricultural subsidies, and have important research significance for promoting the level of automation in my country's agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a structural schematic diagram of a combine harvester of the present invention;

[0016] Figure 2 is a flow chart of image processing of the present invention;

[0017] Figure 3 Schematic diagram of the object distance D of the present invention.

[0018] Figure numerals: 1. Combine harvester, 2. Laser, 3. Industrial camera, 4. Cutting table. DETAILED DESCRIPTION

[0020] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] The embodiment of the present invention provides a method and a device for detecting the cutting width of a combine harvester.

[0022] The present invention provides a combined harvester cutting width detection device, which includes a laser 2, an industrial camera 3, a Hall sensor, an embedded processor and a display device. The laser 2 is arranged above the header of the combine harvester, and the industrial camera 3 is installed above the header 4 of the combine harvester and above the laser 2. The industrial camera 3, the Hall sensor and the display device are all electrically connected to the embedded processor, and the embedded processor is used for data processing, analysis and storage. The Hall sensor is installed at a position where the hub speed can be accurately measured, and a permanent magnet is installed on the driving wheel to cooperate with the Hall sensor. The magnet should be fixed on the horizontal plane between the driving wheel hydraulic motor and the gear to ensure that the Hall element can pass through when the driving wheel rotates. The Hall sensor is connected to the data acquisition card to ensure the stability of data transmission. A data acquisition card suitable for vehicle speed measurement is selected, and then a wireless signal transmitter is installed and connected to the data acquisition card to transmit the collected vehicle speed data to the embedded processor. The embedded processor is connected to the industrial camera 3 via Ethernet. The embedded processor can also be connected to a touch-sensitive LCD screen.

[0023] The laser 2 performs cyclic detection of objects in the fan-shaped area in a line scanning manner, and uses a mechanical rotating platform to scan. The laser 2 is installed on a rotating platform, and the motor drives the platform to rotate to achieve scanning. The motor shaft is directly connected to the rotating platform. When the motor rotates, the platform and the laser 2 installed on the platform are driven to rotate together. The speed and range of the laser scanning can be adjusted by controlling the speed and rotation angle of the motor.

[0024] The present invention provides a method for detecting the cutting width of a combine harvester, using the above-mentioned detection device for detection, and includes the following steps: Step S1: The laser 2 provided on the combine harvester 1 emits laser light to perform a cyclic scanning of crops in front of the harvesting platform 4; Step S2: The industrial camera 3 provided on the combine harvester 1 acquires an image of the area in front of the harvesting table 4; Step S3: The image acquired by the industrial camera 3 is processed to acquire two straight lines corresponding to the edges of the cut width; Step S4: Based on the two straight lines corresponding to the edges of the cut width acquired, the pixel distance between the two straight lines is calculated, and the pixel distance is the pixel cut width d1; Step S5: Obtain the focal length f and object distance D of the industrial camera 3. The object distance D is the distance between the installation position of the industrial camera 3 and the center of the scanning area of ​​the laser 2. Define the actual cutting width as d2. According to the principle of similar triangles, it can be known that: The actual cutting width is Step S6: The speed of the combine harvester 1 is obtained by a Hall sensor disposed on the combine harvester 1, and the cutting area of ​​the combine harvester 1 is calculated according to the obtained actual cutting width d2 and the speed.

[0025] The area scanned cyclically by the laser 2 in step S1 is a sector-shaped area, the orthographic projection of the harvesting platform 4 in the sector-shaped area falls within the sector-shaped area, and the laser emitted by the laser 2 is blue light. The laser 2 emits laser light to the crops in front of the harvesting platform 4 of the combine harvester 1.

[0026] In step S3, the image acquired from the industrial camera 3 is transmitted to the embedded processor for subsequent processing. The processing of the image acquired by the industrial camera 3 includes the following steps: denoising the image by means of median filtering, then color threshold segmenting the image by using the difference between the blue light emitted by the laser emitter and the crops and the ground, and then feature extraction of the extracted laser irradiation area, that is, edge detection, and processing the image by the Canny edge detection algorithm to highlight the edge of the actual cutting width in the laser irradiation area in the image, wherein the Canny operator first smoothes the image by Gaussian filtering to reduce the influence of noise, and then calculates the gradient amplitude and direction of the image, and accurately detects the edge of the laser irradiation area by steps such as non-maximum suppression and double threshold detection. After detecting the edge of the laser irradiation area, since the laser emission is fan-shaped, its edge portion appears as a straight line segment in the image. Then morphological processing is performed, and the overly thin edge lines are appropriately thickened by using the expansion operation to make them more stable and not easy to break or lose in subsequent processing, so as to better serve the subsequent straight line detection and width calculation operations. Then, straight line detection is performed, and two straight lines corresponding to the edge of the cut width are detected from the edge image using the Hough transform straight line detection method. That is, the straight line representing the edge of the laser irradiation area is detected from the edge image using the Hough transform straight line detection method, and the dividing point between the crop and the ground is found, thereby obtaining the dividing straight line, that is, the straight line corresponding to the edge of the cut width. Among them, the median filter method is used for denoising to make the image smoother and retain the edge details of the image, so as to facilitate the subsequent accurate extraction of the laser irradiation area and related features.

[0027] The color of the laser emitted by the laser transmitter is blue light, which will produce a specific blue area in the image. The color threshold segmentation method can be used to set a threshold according to the blue color range, and the pixels in the image within the threshold range can be extracted. These pixels are the areas irradiated by the laser. In the RGB color space, the value of the blue channel will be relatively high. The threshold range of the blue channel can be set to extract the laser irradiation area. For example, for an 8-bit image (channel value range is 0-255), if the value of the blue channel is greater than 180, and the values ​​of the red and green channels are relatively low (such as less than 80), the pixel is judged to belong to the laser irradiation area.

[0028] Specifically, two straight lines are defined as the first straight line and the second straight line. The polar coordinate parameters of the first straight line and the second straight line in the image are detected by using the Hough transform straight line detection method as (ρ1, θ1) and (ρ2, θ2), respectively. The polar coordinate equations of the first straight line and the second straight line are converted into rectangular coordinate equations. The polar coordinate equation of the first straight line is ρ1 = x1 cosθ1 + y1 sinθ1, then the rectangular coordinate equation of the first straight line is The polar coordinate equation of the second straight line is ρ2 = x2cosθ2 + y2 sinθ2, so the rectangular coordinate equation of the second straight line is According to the rectangular coordinate equation of the first straight line, select the intercept point P1 (x1, y1) of the first straight line in the rectangular coordinate system, and according to the rectangular coordinate equation of the second straight line, select the intercept point P2 (x2, y2) of the second straight line in the rectangular coordinate system. Use the distance formula between the two points To calculate the distance between the two points, we can get the pixel distance between the two lines.

[0029] In this embodiment, the intercept point P1 (x1, y1) is the cross-intercept point, that is, the coordinates of the intersection of the rectangular coordinate equation of the first straight line with the X-axis in the rectangular coordinate system, and P2 (x2, y2) is the cross-intercept point, that is, the coordinates of the intersection of the rectangular coordinate equation of the second straight line with the X-axis in the rectangular coordinate system.

[0030] In step S5, the step of obtaining the focal length f of the industrial camera 3 includes using a chessboard calibration plate. The chessboard is placed in the camera field of view, and multiple sets of images at different angles and positions are taken. The focal length f of the industrial camera 3 can be calculated by using the pixel coordinates of the corner points of the chessboard in the image through a specific algorithm (such as the Zhang Zhengyou calibration method).

[0031] Obtaining the object distance D of the industrial camera 3 in step S5 includes the following steps: obtaining the height H1 of the installation position of the industrial camera 3 from the horizontal ground and the shortest distance H2 between the vertical plane where the industrial camera 3 is located and the vertical plane where the laser emitter hits the crop, and using the Pythagorean theorem of triangles to obtain the object distance

[0032] In step S6, calculating the cutting swath area of ​​the combine harvester 1 includes the following steps: According to the acquired vehicle speed, the moving distance of the combine harvester 1 within a certain period of time can be obtained, and the cutting swath area can be obtained by multiplying the moving distance by the actual cutting swath width d2.

[0033] Of course, the present invention is not limited to the above-described embodiments. Several other embodiments based on the design concept of the present invention are also provided below.

[0034] For example, in other embodiments, different from the embodiments described above, the intercept point P1 (x1, y1) can also be the vertical intercept point, that is, the coordinates of the intersection of the rectangular coordinate equation of the first straight line with the Y-axis in the rectangular coordinate system, and P2 (x2, y2) can also be the vertical intercept point, that is, the coordinates of the intersection of the rectangular coordinate equation of the first straight line with the Y-axis in the rectangular coordinate system.

[0035] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A method for detecting the cutting width of a combine harvester, characterized in that: The following steps are involved: Step S1: a laser device installed on the combine harvester emits laser light to perform a cyclic scan on the crops in front of the harvesting platform; Step S2: continuously acquiring images of the area in front of the harvesting platform through an industrial camera installed on the combine harvester; Step S3: Process the image acquired by the industrial camera to obtain two straight lines corresponding to the edges of the cut width; Step S4: according to the two straight lines corresponding to the acquired cut width edge, the pixel distance between the two straight lines is solved, and the pixel distance is the pixel cut width d1; Step S5: Obtain the focal length f and object distance D of the industrial camera. The object distance D is the distance between the installation position of the industrial camera and the center of the area scanned by the laser. Define the actual cutting width as d2. According to the principle of similar triangles: The actual cutting width is Step S6: The speed of the combine harvester is obtained by a Hall sensor arranged on the combine harvester, and the cutting area of ​​the combine harvester is calculated according to the obtained actual cutting width d2 and the speed.

2. A method for detecting the cutting width of a combine harvester according to claim 1, characterized in that: The area scanned cyclically by the laser is a fan-shaped area, and the laser emitted by the laser is blue light.

3. A method for detecting the cutting width of a combine harvester according to claim 2, characterized in that: In step S3, the processing of the image acquired by the industrial camera includes the following steps: denoising the image by means of median filtering, then color threshold segmenting the image by using the difference between the blue light emitted by the laser transmitter and the crops and the ground, and then feature extraction of the extracted laser irradiation area, that is, edge detection, and processing the image by the Canny edge detection algorithm to highlight the edge of the cutting platform and the actual cutting width in the laser irradiation area of ​​the image, using morphological processing to appropriately thicken the overly thin edge lines, and then performing straight line detection, using the Hough transform straight line detection method to detect two straight lines corresponding to the cutting width edges from the edge image.

4. The method for detecting the cutting width of a combine harvester according to claim 1, characterized in that: The two straight lines are the first straight line and the second straight line. The polar coordinate parameters of the first straight line and the second straight line in the image are detected by the Hough transform straight line detection method as (ρ1, θ1) and (ρ2, θ2) respectively. The polar coordinate equations of the first straight line and the second straight line are converted into rectangular coordinate equations. The polar coordinate equation of the first straight line is ρ1 = x1 cosθ1 + y1 sinθ1, so the rectangular coordinate equation of the first straight line is The polar coordinate equation of the second straight line is ρ2 = x2cosθ2 + y2 sinθ2, so the rectangular coordinate equation of the second straight line is According to the rectangular coordinate equation of the first straight line, select the intercept point P1 (x1, y1) of the first straight line in the rectangular coordinate system, and according to the rectangular coordinate equation of the second straight line, select the intercept point P2 (x2, y2) of the second straight line in the rectangular coordinate system. Use the distance formula between the two points To calculate the distance between the two points, we can get the pixel distance between the two lines.

5. The method for detecting the cutting width of a combine harvester according to claim 4, characterized in that: The intercept point P1(x1,y1) is the horizontal intercept point, and P2(x2,y2) is the horizontal intercept point.

6. A method for detecting the cutting width of a combine harvester according to claim 4, characterized in that: The intercept point P1(x1,y1) is the vertical intercept point, and P2(x2,y2) is the vertical intercept point.

7. The method for detecting the cutting width of a combine harvester according to claim 1, characterized in that: Obtaining the object distance D of the industrial camera in step S5 includes the following steps: obtaining the height H1 of the installation position of the industrial camera from the horizontal ground and the shortest distance H2 between the vertical plane where the industrial camera is located and the vertical plane where the laser emitter hits the crop, and using the Pythagorean theorem of triangles to obtain the object distance 8. The method for detecting the cutting width of a combine harvester according to claim 1, characterized in that: In step S6, calculating the cutting swath area of ​​the combine harvester includes the following steps: According to the acquired vehicle speed, the moving distance of the combine harvester within a certain period of time can be obtained, and the cutting swath area can be obtained by multiplying the moving distance by the actual cutting swath width d2.

Images