Speed measuring device and speed measuring method suitable for tractor field operation

Abstract

The invention relates to the technical field of target speed measurement, in particular to a speed measurement device and method suitable for tractor field operation, and the speed measurement device comprises binocular camera equipment and a gyroscope; the speed measurement method comprises the steps that images shot by the binocular imaging device and angles measured by the gyroscope are transmitted to the processor to be processed, the processor conducts image analysis on the image overlapping parts of the paired images through an image matching algorithm, image coordinates of image feature points are obtained, and the relative actual distance of tractor operation is obtained through conversion calculation; the interval time shot by the binocular imaging device, shot images and angle change information, measured by the gyroscope, of the whole tractor are transmitted to the processor to be processed, and the real-time speed of the tractor when the tractor works on the road surface is obtained. The method can measure the speed of the tractor according to the actual working condition of the tractor, has the advantages of high precision, good real-time performance and the like, and can measure the real-time speed of the tractor when the tractor works on an uneven road surface.

Description

Technical Field

[0001] The invention relates to the technical field of target speed measurement, and in particular to a speed measurement device and a speed measurement method suitable for tractor field operation. Background Art

[0002] As an important agricultural machinery, tractors often operate under variable and complex working road conditions, which directly affect the performance of tractors. Especially on uneven roads, wheels are prone to slip, which significantly reduces work efficiency and economy. Therefore, controlling the slip rate of tractors near the optimal slip rate has become the key to solving the problem. The key to determining the slip rate lies in the accurate measurement of the real-time speed of the tractor.

[0003] In the process of measuring the real-time speed of the tractor, existing speed measurement technologies, such as GPS speed measurement system, wheel angular velocity sensor, Doppler observation value speed measurement and carrier differential speed measurement, will cause great interference to the measurement due to the vibration, noise and tire slip generated by the tractor during driving, greatly reducing the accuracy of speed measurement and making it difficult to meet the stringent requirements for real-time speed measurement with controlled slip rate.

[0004] The binocular imaging device uses two parallel imaging devices to quickly shoot the objects to be photographed at the same time interval. In a short time interval, there is an overlapping part in the images taken by the two imaging devices. The image matching technology is used to calculate the pixel offset between the image pairs, and the vehicle's own movement speed is obtained by combining the calibration parameters and interval time. However, when the tractor is operating on an uneven road, the body of the vehicle will experience large ups and downs, which will cause the image pair to have a directional deviation, reducing the accuracy of the speed measurement.

[0005] In summary, obtaining the real-time speed of the tractor has always been a key technical bottleneck restricting the control of the tractor's slip rate. In view of this problem, it is urgent to seek a method that can measure the real-time speed of the tractor when the tractor body has large fluctuations. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a speed measuring device and a speed measuring method suitable for tractor field operations. For the tractor operating in the field, real-time speed measurement can be achieved, and the actual distance traveled by the tractor due to the ups and downs of the tractor body can be accurately calculated during the speed measurement process, thereby greatly improving the accuracy of speed measurement.

[0007] The purpose of the present invention is to achieve the following technical solution. According to the present invention, a speed measuring device suitable for tractor field operation includes a binocular camera device for setting on the side of the tractor body and a gyroscope for setting at the center of the tractor longitudinal beam; the binocular camera device includes an imaging device I and an imaging device II, the imaging device I and the imaging device II are horizontally installed on the side of the tractor body when measuring speed, the optical axis of the imaging device is perpendicular to the running direction of the tractor and faces the outside of the tractor body when measuring speed, and the imaging device I and the imaging device II are connected to the same clock signal source when measuring speed.

[0008] A speed measurement method suitable for tractor field operation comprises the following steps:

[0009] The binocular imaging device is horizontally mounted on the side of the tractor body, and the optical axis of the binocular imaging device is perpendicular to the crops beside the body. When the tractor is working in the field, the binocular imaging device can clearly capture the crops beside the tractor body, and use the crops beside the body as an imaging collection reference for measuring the vehicle speed;

[0010] Connecting two imaging devices in the binocular imaging device to the same clock signal source, setting the shooting time and time interval of the two imaging devices in the binocular imaging device, so that the shooting time intervals of the two imaging devices when starting shooting are consistent, and the images captured by the two imaging devices can correspond to each other;

[0011] The gyroscope is installed at the center of the tractor longitudinal beam to measure the angle change of the tractor body when it is working on uneven roads, and the angle between the actual running displacement of the tractor and the horizontal displacement is obtained;

[0012] The images taken by the binocular imaging device and the angles measured by the gyroscope are transmitted to the processor for processing. The processor uses an image matching algorithm to analyze the overlapping parts of the paired images to obtain the image coordinates of the image feature points. The relative actual distance of the tractor operation is obtained through transformation and calculation.

[0013] The interval time taken by the binocular imaging device, the taken images and the angle change information of the whole vehicle measured by the gyroscope are transmitted to the processor for processing to obtain the real-time speed of the tractor when working on the road.

[0014] Furthermore, when measuring speed, the gyroscope's rotation axis is parallel to the axis that the tractor rotates when traveling on uneven terrain.

[0015] Furthermore, the crop picture captured by the imaging device I is image A, and when the tractor is running on the horizontal ground, the crop picture captured by the imaging device II in the same time interval is image B;

[0016] The coordinates of a feature point in image A at time t1 are (x1, y1), and the coordinates of the same feature point in image B at time t2 are (x2, y1). According to the transformation function of the imaging system, the above image coordinates are converted into ground coordinates, which are (X1, Y1) and (X2, Y1) respectively. Then the moving distance of the crop relative to the tractor is S1 = Y2-Y1.

[0017] The tractor moves the same distance, i.e. S1, and the relative movement direction is opposite to that of the crops.

[0018] Furthermore, the crop image captured by the imaging device I is image A, and when the tractor body is running on the uneven ground, the crop image captured by the imaging device II in the same time interval is image C;

[0019] The overlapping area of ​​image A and image C is matched and analyzed. Assume that the overlapping area for analysis is area D. The coordinates of a feature point of image A in area D at time t1 are (x1, y1), and the coordinates of the same feature point in image C in area D at time t2 are (x2, y2). The image coordinates are converted into ground coordinates, which are (X1, Y1) and (X2, Y2) respectively. It can be obtained that the crops move from (X1, Y1) to (X2, Y2) relative to the tractor. The distance traveled by the tractor is equal to the moving distance of the crops, and the relative movement directions of the two are opposite. The gyroscope directly measures the actual undulation angle of the tractor body to obtain the real-time vehicle undulation angle θ. Therefore, the actual movement distance of the tractor is:

[0020]

[0021] or

[0022]

[0023] Furthermore, when calculating the vehicle speed, the time difference of the frame synchronization signal output by the binocular imaging device when taking images is detected to obtain the shooting interval time. The real-time vehicle speed required by the present invention is obtained by combining the measurement information of the two devices through the processor. The acquisition time interval of the imaging device is:

[0024] Δt=t2-t1

[0025] The real-time speed of the vehicle is:

[0026]

[0027] or

[0028]

[0029] Compared with the prior art, the invention is beneficial in that: compared with the traditional GPS speed measurement system, the method can measure the speed of the tractor according to the actual operating conditions of the tractor, has the advantages of high accuracy and good real-time performance, and can realize the real-time speed measurement of the tractor when the body of the tractor undergoes large ups and downs when operating on uneven roads.

[0030] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the following specifically cites a preferred embodiment and describes it in detail with the accompanying drawings as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is an implementation flow chart of an embodiment of a speed measurement method applicable to tractor field operation of the present invention;

[0032] Figure 2 A schematic structural diagram of a tractor installed with a speed measuring device for tractor field operation according to an embodiment of the present invention;

[0033] Figure 3 A simplified structural diagram of a tractor chassis installed with an embodiment of a speed measuring device suitable for tractor field operation according to the present invention;

[0034] Figure 4 An image A acquired by an imaging device I;

[0035] Figure 5 Image B acquired by imaging device II;

[0036] Figure 6 Image C acquired by imaging device II;

[0037] Figure 7 for Figure 2 The schematic diagram of displacement change of the image collected by the binocular imaging device when the tractor is traveling on a relatively horizontal road is shown;

[0038] Figure 8 for Figure 2 The diagram shown is a schematic diagram of the displacement change of the image collected by the binocular imaging device when the tractor is driving on an uneven road;

[0039] Fig. 9 Schematic diagram of the distance measurement principle of two images collected by binocular imaging equipment in the coordinate system.

[0040] [Reference Signs]

[0041] 1. Engine cylinder head; 2. Tractor front wheel; 3. Exhaust pipe; 4. Hood; 5. Cab; 6. Rear wheel fender; 7. Front axle; 8. Engine; 9. Longitudinal beam; 10. Rear axle. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0043] An embodiment of a speed measuring device suitable for tractor field operation of the present invention is as follows: Figure 2 , Figure 3 As shown, the speed measuring device includes a binocular imaging device and a gyroscope. The binocular imaging device and the gyroscope are installed on the tractor body and used in combination. The binocular imaging device is used to photograph the crops beside the tractor body when the tractor is working, and the gyroscope is used to measure the angle of the tractor body when it rises and falls. The data measured by the two devices are integrated to obtain the real-time measurement of the tractor speed on the uneven road surface.

[0044] In this embodiment, the binocular imaging device is installed in the middle and lower part of the side of the tractor body, which is convenient for photographing the crops beside the tractor when it is working. Figure 2 As shown, the binocular imaging device is installed at the hood 4 on the side of the tractor body. The binocular imaging device includes an imaging device I and an imaging device II, and the two imaging devices are horizontally installed on the side of the tractor body. The optical axis of the imaging device is perpendicular to the running direction of the tractor. The two imaging devices are connected to the same clock signal source, so that the shooting time and shooting interval of the two imaging devices can be synchronized.

[0045] The gyroscope is installed at the center of the tractor longitudinal beam to facilitate the measurement of the angle change of the tractor body. Figure 3 As shown, the gyroscope is installed at the center of the tractor longitudinal beam 3.

[0046] An embodiment of the speed measurement method of the present invention is applicable to tractor field operation, as shown in FIG. Figure 1 As shown, the data results measured by the binocular imaging device and the gyroscope are fused to obtain the real-time operating speed of the tractor in the field, which specifically includes the following steps:

[0047] 1. Install the binocular imaging device horizontally on the side of the tractor body, and make the optical axis of the binocular imaging device perpendicular to the crops beside the body, that is, perpendicular to the running direction of the tractor. When the tractor is working in the field, the two imaging devices will clearly capture the crops beside the tractor body.

[0048] According to the actual operating conditions of the tractor, when the tractor is running and operating, the body vibrates violently. In order to reduce the measurement error caused by the tractor's own vibration, the binocular imaging device is installed horizontally on the tractor body to reduce the measurement error caused by the imaging device's own structure, thereby improving the measurement accuracy.

[0049] According to the actual operating conditions of tractors, especially in the Central Plains of my country, the terrain is flat and the land area is vast, which is suitable for large-scale mechanized operations. If the binocular imaging device is installed on a small road outside the field or on a distant fixed object such as a large tree in the field, although the body posture changes and movement status of the tractor can be obtained, the imaging range will be too large, which greatly reduces the accurate shooting of the tractor and reference objects. The surrounding environment interferes too much with image matching, increasing the complexity and calculation amount of image matching. Therefore, the imaging device is installed on the tractor body, which can achieve accurate shooting of crops next to the body, accurately match the information changes of each pair of images, and obtain accurate moving vehicle speed.

[0050] According to the actual operating conditions of the tractor, large areas of land in my country are planted with crops such as wheat, corn, and soybeans. When the tractor is operating, crops, as the most direct reference object, have a high degree of recognition and can easily distinguish the same feature points of objects in each frame and obtain the moving distance of the same object. By real-time shooting of the crops on the side of the tractor, the moving distance and movement speed of the crops are analyzed, thereby reflecting the real-time movement speed of the tractor.

[0051] In summary, the present invention selects to install the binocular imaging equipment on the body of the tractor on the side and selects the crops beside the tractor body as the imaging collection reference for measuring the body speed.

[0052] 2. Connect the two imaging devices in the binocular imaging device to the same clock signal source, set the shooting time and time interval of the two imaging devices in the binocular imaging device, so as to accurately control the shooting time of the two imaging devices, make the shooting time interval of the two imaging devices consistent after starting, so that the images captured by the two imaging devices can correspond to each other and match, so as to obtain the relative distance of the tractor operation.

[0053] The present invention needs to obtain accurate real-time speed of the tractor, but because different imaging devices have their own shooting clock signal sources, when the frequencies of the clock signal sources are inconsistent, the frame signal times of different imaging devices will produce relative offsets, which greatly increases the error of matching the pictures shot by the two imaging devices. Therefore, the clock signal sources of the two imaging devices need to be unified before shooting, and the shooting time of the imaging device II is adjusted by setting the delay, so that the shooting interval of the imaging device I and the imaging device II can be determined, and the time interval of each shooting of the two imaging devices after startup is kept consistent, thereby ensuring that the shooting of crops on the same side can be completed within a precise and stable time interval, reducing the adverse effects of setting the binocular imaging device on the speed measurement method.

[0054] 3. Install the gyroscope at the center of the tractor longitudinal beam 3 to measure the angle change of the tractor body when it is working on uneven roads, and obtain the angle between the actual operating displacement and the horizontal displacement of the tractor.

[0055] Traditional gyroscopes are large in size, high in power consumption, easily disturbed by the external environment, have poor stability, are difficult to install, and have large measurement errors, so they are not suitable for accurate measurement of the real-time speed of tractors. According to the requirements of the present invention, a new ADXRS gyroscope is selected, which has high measurement accuracy, small structural size, and is easy to install on the tractor. It greatly reduces the influence of the vibration and noise of the tractor on the measurement angle accuracy during the measurement process, and has strong stability and robustness.

[0056] The present invention uses a gyroscope to measure the tilt angle of the whole vehicle of the tractor. The gyroscope includes a capacitive detection element attached to a harmonic body. When the tractor drives the gyroscope to change its angle, according to the Coriolis effect, the mass element of the harmonic body and the mass element in the middle of the frame structure will compress and stretch the spring element fixed around, thereby generating displacement. The generated displacement will generate a differential capacitance in the capacitive element. The generated differential capacitance is the rotational angular velocity of the object obtained. These capacitive detection elements are all beams made of silicon material. They intersect with two groups of stationary silicon beams attached to a substrate, thereby forming two capacitors with equal nominal values. The generated differential capacitance is the rotational angular velocity of the tractor body. The structure and working principle of the above-mentioned gyroscope are prior art and will not be repeated here.

[0057] The gyroscope used in the present invention can be installed at any angle anywhere on the rotating object, but when measurement is required, the gyroscope's rotation axis is parallel to the axis of rotation of the tractor when it travels on uneven ground, and the measured angle can be obtained by integrating the angular velocity.

[0058] The present invention aims to measure the change in the body angle caused by the wheels of a tractor driving on an uneven road surface. Since the wheel air pressure of the tractor will constantly change when operating under such road conditions, if the gyroscope is installed at the center of the wheel, not only will the installation be difficult, but the center of mass position of the tractor will also be difficult to determine, and the measurement accuracy will be greatly reduced. Therefore, installing the gyroscope at the center of the tractor chassis longitudinal beam can effectively improve the measurement accuracy and reduce the measurement error caused by the wheel air pressure fluctuation.

[0059] In summary, the present application selects a gyroscope and installs it at the center of the longitudinal beam of the tractor chassis.

[0060] 4. The images taken by the binocular imaging device and the data such as the tilt angle measured by the gyroscope are transmitted to the processor for processing. The processor uses the image matching algorithm to analyze the overlapping parts of the paired images (the images taken by imaging device I and imaging device II in sequence within a time interval are a pair of images) to obtain the image coordinates of the image feature points, and the relative actual distance of the tractor's operation is obtained through transformation and calculation.

[0061] In order to analyze the images taken by the binocular imaging device, camera calibration is a key step in computer vision, and its purpose is to determine the intrinsic and extrinsic parameters of the camera. The intrinsic parameters describe the geometric and optical properties of the camera, such as focal length, point coordinates, etc.; the extrinsic parameters describe the position and posture of the camera in the world coordinate system.

[0062] Take the images captured in one time interval as an example for illustration. The crop image captured by the imaging device I is image A. Figure 4 When the tractor is running on a horizontal ground, the crop image captured by the imaging device II is image B, as shown in the attached figure. Figure 5 When the tractor body is running on the uneven ground, the crop picture collected by the imaging device II is image C, as shown in the attached figure. Figure 6 shown.

[0063] The present invention proposes a comparative working condition. When the tractor is operating on a relatively stable road surface, the same object in the image captured by the binocular imaging device has only horizontal displacement and no vertical displacement. Figure 7As shown in the figure, assuming that the coordinates of a feature point in image A at time t1 are (x1, y1), and the coordinates of the same feature point in image B at time t2 are (x2, y1), the above image coordinates need to be converted into ground coordinates to calculate the actual moving distance. The ground coordinate points need to be transformed by the linear imaging model and the nonlinear imaging model during the imaging process. The linear imaging model is an ideal pinhole imaging model, and the nonlinear imaging model is the distortion error generated by the imaging device lens, including eccentric distortion, thin prism distortion, radial distortion, and tangential distortion. According to the transformation function of the imaging system (obtained through the intrinsic and extrinsic parameters of the imaging device), the above image coordinates are converted into ground coordinates, which are (X1, Y1) (X2, Y1), respectively. Then the moving distance of the crop relative to the tractor is

[0064] S1=Y2-Y1

[0065] The tractor moves the same distance, i.e. S1, and the relative movement direction is opposite to that of the crops.

[0066] When a tractor is operating on an uneven road, the body of the tractor will experience large fluctuations. Therefore, the feature points of the same object in the image captured by the imaging device will not only have horizontal displacement but also vertical displacement. Figure 8 As shown. Matching analysis is performed on the overlapping area of ​​image A and image C. Let the overlapping area for analysis be area D. Assume that the coordinates of a feature point in image A at time t1 in area D are (x1, y1), and the coordinates of the same feature point in image C at time t2 in area D are (x2, y2). The image coordinates are converted into ground coordinates, which are (X1, Y1) and (X2, Y2), respectively. From this, it can be obtained that the crop moves from (X1, Y1) to (X2, Y2) relative to the tractor (binocular imaging device). Fig. 9 As shown, the distance traveled by the tractor is equal to the distance moved by the crops, and the relative moving directions of the two are opposite.

[0067] Due to the limitations of lens imaging itself, the limitations of the manufacturing process, and the different shooting angles of the two imaging devices, the imaging process of the crop coordinate points will require the transformation of linear imaging models and nonlinear imaging models. The nonlinear imaging model is the geometric distortion error generated by the imaging device lens, such as radial distortion and tangential distortion. These distortions will directly affect the matching of the actual coordinate system and the image coordinate system, especially the angle has a large error during the transformation process, thereby affecting the matching of the actual coordinate system and the image coordinate system. Therefore, the present invention uses a gyroscope to directly measure the actual body ups and downs angle of the tractor to obtain the real-time vehicle ups and downs angle, such as Fig. 9 The angle represented by θ in the figure can be used to improve the accuracy of measuring the actual driving distance.

[0068] Therefore, the actual movement distance of the tractor is:

[0069]

[0070] or

[0071]

[0072] Before calculating the actual movement distance of the tractor, it is first necessary to match and analyze the paired images through the image matching algorithm, obtain the overlapping parts of the images and the image feature points, and obtain the image coordinates of the image feature points, and then transform them into ground coordinates for calculation. The image matching algorithm is mainly divided into template-based matching, grayscale-based matching, transform domain-based matching and feature-based matching according to different principles.

[0073] Among them, feature-based matching is one of the most commonly used matching algorithms at present. This algorithm extracts feature information sets such as points, lines, and contours in the image, and then matches the feature sets to correspond. While reducing the amount of calculation and improving efficiency, it has a certain robustness to the grayscale changes of the image. However, this algorithm only uses the feature information of a small part of the image, so it has very high requirements for the precision and accuracy of feature extraction and feature matching. Due to the complex changes of crops during tractor field operations, the captured images are degraded by motion blur, and the error rate of feature-based matching methods increases sharply. Template-based matching has the characteristics of simplicity and directness, a large template range, and uses more image information, so it has a certain robustness to image motion blur. However, the template-based matching algorithm needs to extract the template in one image in advance, and then perform a series of sliding matches in another image. The extraction and matching are repeated, the amount of calculation is large, and the matching speed is slow, which cannot meet the real-time requirements of this application.

[0074] To sum up, in order to make full use of the advantages of these two algorithms and overcome their disadvantages, the present invention utilizes the above-mentioned feature-based matching method and template-based matching method for collaborative matching, utilizes the feature matching algorithm to provide a basis for the template matching algorithm, improves the matching speed, and utilizes the template matching algorithm to improve the matching accuracy. This is the prior art and will not be repeated here. The image matching algorithm includes the SURF matching algorithm.

[0075] 5. The interval time taken by the binocular imaging device, the taken images and the angle change information of the whole vehicle measured by the gyroscope are transmitted to the processor for processing to obtain the real-time speed of the tractor when operating on uneven roads.

[0076] According to the above-mentioned image matching analysis, the displacement change of the same feature point in the two images is obtained, and the displacement change of the same crop feature point in the image is directly obtained. When the tractor encounters a road subsidence (or road protrusion) when driving on an uneven road, the tractor body will suddenly undergo a downward (or upward) displacement change, but in the picture taken by the binocular imaging device, the same crop feature point is displaced upward (or downward). The displacement change of the feature point reflects the actual real-time running distance of the tractor. The driving direction of the tractor is opposite to the direction shown by the crop feature point in the image. The actual undulation angle of the body is directly measured by a gyroscope. At the same time, the time difference of the frame synchronization signal output by the binocular imaging device when shooting the image is detected to obtain the shooting interval time, and the real-time vehicle speed required by the present invention is obtained by combining the measurement information of the two devices through the processor.

[0077] The acquisition time interval of the imaging device is:

[0078] Δt=t2-t1

[0079] The real-time speed of the vehicle is:

[0080]

[0081] or

[0082]

[0083] The speed measurement method of the present invention is summarized as follows: when measuring the speed of a tractor, the present invention firstly uses binocular imaging equipment to continuously and in real time photograph the crops beside the tractor body at the same time interval to obtain multiple pairs of overlapping image information; secondly, the image matching technology is used to calculate the relative movement distance of the image, but because the tractor's operating conditions have uneven roads, the measured running distance and the actual running distance are deviated, which affects the accuracy of speed measurement. Therefore, a gyroscope is used to measure the angle, and the measured data is combined and analyzed to obtain the actual running distance of the tractor; finally, the average speed of the tractor in a short period of time is obtained by the processor, and the average speed is used instead of the real-time speed. Compared with the traditional GPS speed measurement system, the present method can measure the speed of the tractor according to the actual operating conditions of the tractor, and has the advantages of high accuracy and good real-time performance, and can realize the measurement of the real-time speed of the tractor when the body of the tractor is operating on uneven roads and has large fluctuations.

[0084] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A speed measuring device suitable for tractor field operation, characterized in that: It includes a binocular camera device for being arranged on the side of a tractor body and a gyroscope for being arranged at the center position of a tractor longitudinal beam; the binocular camera device includes an imaging device I and an imaging device II. The imaging device I and the imaging device II are horizontally installed on the side of the tractor body when measuring speed. When measuring speed, the optical axis of the imaging device is perpendicular to the running direction of the tractor and faces the outside of the tractor body. The imaging device I and the imaging device II are connected to the same clock signal source when measuring speed.

2. A speed measurement method suitable for tractor field operation, characterized in that: The following steps are involved: The binocular imaging device is horizontally mounted on the side of the tractor body, and the optical axis of the binocular imaging device is perpendicular to the crops beside the body. When the tractor is working in the field, the binocular imaging device can clearly capture the crops beside the tractor body, and use the crops beside the body as an imaging collection reference for measuring the vehicle speed; Connecting two imaging devices in the binocular imaging device to the same clock signal source, setting the shooting time and time interval of the two imaging devices in the binocular imaging device, so that the shooting time intervals of the two imaging devices when starting shooting are consistent, and the images captured by the two imaging devices can correspond to each other; The gyroscope is installed at the center of the tractor longitudinal beam to measure the angle change of the tractor body when it is working on uneven roads, and the angle between the actual running displacement of the tractor and the horizontal displacement is obtained; The images taken by the binocular imaging device and the angles measured by the gyroscope are transmitted to the processor for processing. The processor uses an image matching algorithm to analyze the overlapping parts of the paired images to obtain the image coordinates of the image feature points. The relative actual distance of the tractor operation is obtained through transformation and calculation. The interval time taken by the binocular imaging device, the taken images and the angle change information of the whole vehicle measured by the gyroscope are transmitted to the processor for processing to obtain the real-time speed of the tractor when working on the road.

3. A speed measurement method suitable for tractor field operation according to claim 2, characterized in that: When measuring speed, the gyroscope's rotation axis is parallel to the axis that the tractor rotates when driving on uneven ground.

4. A speed measurement method suitable for tractor field operation according to claim 2, characterized in that: The crop image captured by imaging device I is image A. When the tractor is running on the horizontal ground, the crop image captured by imaging device II in the same time interval is image B. The coordinates of a feature point in image A at time t1 are (x1, y1), and the coordinates of the same feature point in image B at time t2 are (x2, y1). According to the transformation function of the imaging system, the above image coordinates are converted into ground coordinates, which are (X1, Y1) and (X2, Y1) respectively. Then the moving distance of the crop relative to the tractor is S1=Y2-Y1 The tractor moves the same distance, i.e. S1, and the relative movement direction is opposite to that of the crops.

5. The speed measurement method for tractor field operation according to claim 2, characterized in that: The crop image captured by imaging device I is image A. When the tractor body is running on the uneven ground, the crop image captured by imaging device II in the same time interval is image C. The overlapping area of ​​image A and image C is matched and analyzed. Assume that the overlapping area for analysis is area D. The coordinates of a feature point of image A in area D at time t1 are (x1, y1), and the coordinates of the same feature point in image C in area D at time t2 are (x2, y2). The image coordinates are converted into ground coordinates, which are (X1, Y1) and (X2, Y2) respectively. It can be obtained that the crops move from (X1, Y1) to (X2, Y2) relative to the tractor. The distance traveled by the tractor is equal to the moving distance of the crops, and the relative movement directions of the two are opposite. The gyroscope directly measures the actual undulation angle of the tractor body to obtain the real-time vehicle undulation angle θ. Therefore, the actual movement distance of the tractor is: or 6. A speed measurement method suitable for tractor field operation according to claim 2, characterized in that: When calculating the vehicle speed, the frame synchronization signal time difference output by the binocular imaging device when taking images is detected to obtain the shooting interval time. The real-time vehicle speed required by the present invention is obtained by combining the measurement information of the two devices through the processor. The acquisition time interval of the imaging device is: Δt=t2-t1 The real-time speed of the vehicle is: or

Images