| The regular harvesting(stumping)of Caragana korshinskii is an important technical means for the rejuvenation and regeneration of Caragana korshinskii.At present,the harvesting(stumping)of Caragana korshinskii is still done by manually carrying the brush cutter,which is low in efficiency and high in work intensity.Since the stubble height is required to be 10cm after cutting,large-scale harvesting machinery cannot complete the harvesting(stumping)of Caragana korshinskii with complex growth environment and terrain because it does not have the terrain adaptive function.Thus,this paper takes Caragana korshinskii as the research object,researches the biological characteristics of Caragana korshinskii and the agronomic requirements of Caragana korshinskii,and develops the target recognition and positioning technology for intelligent harvesting(stumping)Caragana korshinskii.This article analyzes the biological characteristics and topography of Caragana korshinski and the agronomic requirements of the harvesting(stumping)of Caragana korshinskii,and clarifies the performance indicators of the Caragana harvesting(keeping stubble at a certain height after cutting)equipment.It designs overall plan of the vision system for positioning,and comprehensively considers the technical requirements and performance of the vision system,completes the selection of related hardware.It carries out the research into Caragana korshinski target recognition technology.It realizes the Caragana korshinski target by comparing different image gray-scale methods,image segmentation,image filtering,morphological operations,repair and fitting of korshinskii target,and it realizes the extraction of its skeleton by refining the korshinskii.Based on the target recognition technology of Caragana korshinskii,it completes the monocular vision system design and ranging technology research.The system acquires two images of Caragana korshinskii at different locations,and realizes the location of the root neck of Caragana korshinskii by matching the characteristic points of the root neck of the image.It carries out theoretical analysis,system calibration and indoor and outdoor distance measurement experiments on the distance measurement system,it verifies the performance of the monocular vision system.It designs and researches the laser active vision ranging system.According to its measurement principle,the relative installation position of the camera and the laser is set by comparison,which improves the accuracy of the system and reduces the volume of the system.By establishing the relationship between the pixel coordinates of the laser stripe and the three-dimensional coordinates of the space,it calibrates the system and carries out a ranging experiment to verify the performance of the system.Based on the installation position of the vision system on the cutting table of the Caragana korshinskii stubble machine,it can obtain the information of the topography on both sides of the Caragana korshinskii;based on the topographic features of the root neck of the Caragana korshinskii,it can obtain the complete distance information of the topography of the Caragana korshinskii through interpolation and curve fitting,the root neck of Caragana korshinskii is located indirectly.It realizes the extraction of laser stripe center through image graying,gamma transformation,image segmentation,morphological operation,and Steger algorithm based on Hessian matrix.It carries out laser fringe extraction experiments in different weather environments.In a relatively large illuminance environment,the light can reduce the signal-to-noise ratio of the laser fringe image,and even eliminate the laser fringe;the use of shadow area method and filters can enhance the laser fringe brightness,improve the laser stripe extraction effect.In the north of Yanchi County,Ningxia,it carries out experiments on the root neck positioning of Caragana korshinskii with the monocular vision ranging system and the laser active vision ranging system,the maximum error measured by the monocular vision system is 9.0633mm,the maximum relative error is 1.947%,and the calculation time is 283ms;the maximum error of the laser active vision ranging system is 7mm,the maximum relative error is 2.349%,and the calculation time is 97ms,through the analysis of the experimental results,the feasibility of the ranging system is verified.It analyzes the causes of errors in the ranging system and the effects of the errors on the stubble of Caragana korshinskii.It is based on the Lucky-Richardson algorithm to realize the restoration of the motion blurred image through the point spread function on the image motion blur generated during the experiment. |