Study On Soil Surface Roughness Detection Method Based On Line Structured Light Sensor

Soil surface roughness is an operational parameter that must be considered in field management such as crop planting,irrigation and harvesting,and is one of the key issues to be solved urgently in the development of precision agriculture.Aiming at the problem of lack of rapid detection of soil surface roughness,a method for detecting soil surface roughness based on line structure light sensor is proposed.The semiconductor red laser,industrial camera,computer and brackets are used to construct the image acquisition system.The measurement system is used to obtain images data of the soil surface after calibration.The 3D point cloud data are obtained by MATLAB,and then the 3D reconstruction of the soil surface and the calculation of the surface area are completed.The ratio of the surface area to the plane area of the soil surface is defined as the characterization of the surface roughness of the soil,and the surface roughness of the soil is detected and judged.The main research findings are as follows:(1)Construction and calibration of image acquisition systemA soil surface roughness detection system based on line structure light sensor was built.Among them,the line structure optical laser uses a semiconductor red laser,and the camera is a CMOS industrial camera.The calibration system is calibrated by using a self-made calibration target,and the coordinate image coordinates of the target and the coordinate information in the light plane are obtained.The penalty function constraint method can be used to solve the structural parameters of the line structured light sensor.(2)Research on image processing method of soil surfaceThe soil surface image processing method was determined by comparing different image processing methods and according to the image characteristics of the soil surface line structured light acquired by the system.That is,using the region growing method for image segmentation.The morphological processing order of the image is determined by first performing image expansion and then performing image corrosion.Image edge detection is performed using Canny edge detection method.The gray center of gravity method is used to extract the center of the light bar of the line structure light and the piecewise cubic Hermite interpolation method is used to interpolate the break point in the image.The Delaunay triangulation method is used to realize the three-dimensional reconstruction of the soil surface,and the ratio of the surface area of the soil surface to its corresponding plane area is used to characterize the surface roughness of the soil.The image acquisition ability of the system was verified and the image processing flow and methods were optimized by using four different granularity preparation soil samples.The threshold value of the roughness is determined according to the above-mentioned sample roughness detection result.When G-1 is equal to 0,the soil surface is completely flat.When the absolute value of G-1 is not more than 0.5,the surface of the soil is finer.When the absolute value of G-1 is greater than 0.5 and not more than 1,the surface of the soil is rough.When the absolute value of G-1 is greater than 1,the soil surface is rough.(3)Indoor soil trough verification test results and analysisIn the indoor soil trough,the uncultivated samples and the shallow rotary tillage samples and the deep rotary tillage samples after the rotary tiller operation were selected.Using the image processing methods described above to extract the surface granularity information of the samples in the indoor soil trough and realize the three-dimensional reconstruction of the soil surface.After calculating the surface area of the soil surface particles of each sample the roughness of the sample surface of the indoor soil trough is judged according to the threshold value of the roughness.The result of discriminating the surface roughness of the sample soil is that the uncultivated sample are more dense,the deep rotary tillage samples are more rough and rough,and the shallow rotary tillage samples roughness are between the two.The result conforms to the actual state of the sample.And the measurement result is same as the roughness result measured by the probe method.The discriminant accuracy rate for the three samples was 100%.It is proved that the method can be used to detect and judge the degree of grain roughness on the soil surface.(4)Outdoor soil trough verification test results and analysisIn the outdoor soil trough,the uncultivated samples and the shallow rotary tillage samples and the deep rotary tillage samples after the rotary tiller operation were selected,and the aforementioned methods were used to extract the surface granularity information of the samples in the outdoor soil trough and realize the three-dimensional reconstruction of the soil surface.After calculating the value of the soil surface roughness characteristic amount of samples,the roughness of the samples surface levels were discriminated according to the threshold value of the roughness,and the result was compared with the roughness result measured by the probe method,and the measurement results were the same.Due to the large soil particles in the outdoor soil trough,the surface roughness of the sample soil is judged as the uncultivated samples are more rough,and the shallow tillage samples and the deep tillage samples are rough.However,the calculated values of roughness of uncultivated samples,shallow tillage samples and deep tillage samples increased in turn,in line with the actual state of the sample.It is proved that the method of the study can be used to detect and judge the grain roughness of the surface of the outdoor soil trough.The above research results show that the soil surface roughness detection method based on line structure light sensor can provide a low-cost and easy-to-operate implementation scheme for the rapid and accurate determination of soil surface roughness.