Design And Research Of Fresh Shell Lotus Sorting Machine Based On Machine Vision

The peeling of fresh lotus seeds is an important process in the processing of fresh lotus seeds,which directly affects the processing quality of fresh lotus seeds.Due to the influence of the growing environment,the same lotus pod may contain different degrees of maturity.In order to improve the mechanical breaking quality of fresh shell lotus,lotus seed processing enterprises usually need to manually sort shell lotus of different maturity before shell breaking,so as to use shell breaking equipment with different process parameters for sorting and processing.Due to the low efficiency of manual screening of shell lotus,with the reduction of rural surplus labor,machine sorting of fresh shell lotus is an inevitable requirement for the development of lotus seed processing industry.In order to realize the automation of the sorting of chrysanthemum lotus,this paper adopts the method of combining theoretical analysis and experimental test,with the help of machine vision technology,the online detection of chrysanthemum chrysanthemum with different maturity levels is successfully realized,and the principle prototype development of the fresh chrysanthemum lotus sorting machine is completed.and test.The main work completed by the thesis is as follows:(1)Design of shell lotus image detection system.Tested and analyzed the geometric parameters and color characteristics of the shell lotus at different mature stages,and determined the gray mean value of the G component as the basis for the detection of the mature stage;The selection of the lotus detection hardware system and the design of the shell lotus detection software architecture;combined with the actual detection conditions,the overall scheme layout and design of the image acquisition system are carried out.(2)The design and implementation of the sorting algorithm at the ripening stage of shell lotus.Use the ROI(region of interest)to obtain a specific area of the original image;by analyzing the gray histogram of the image in different color components,select the a component of the Lab color space to perform background segmentation on the image;based on the difference in the threshold segmentation size of shell lotus at different maturity stages,and the adaptive segmentation feature of OTSU(maximum inter-class variance method)is used to segment the background;morphologically process the segmented binary image to eliminate noise and fill holes;in order to completely extract the area where the shell lotus is located,the optimal contour fitting circle method is used for the binary image.The image was contour fitted;the gray mean value of the G component in the RGB color space was obtained by the AND operation,and the detection thresholds were set to 65 and 158,which were used as the basis for distinguishing the wax-ripening stage and the dry-ripening stage.(3)The whole machine structure and control system design.Based on the requirements of automatic sorting of shell lotus,the mechanical structure design of shell lotus sorting machine is completed;in order to facilitate the debugging of the image acquisition system,the detection bracket is set to be adjustable with 6 degrees of freedom,and NX10.0 software is used to carry out three-dimensional construction of the whole machine.According to the sorting control requirements,the prototype control scheme is designed in detail and the components are selected;the control program is designed by multi-thread programming.(4)Prototype manufacturing and testing.The principle prototype manufacturing and the whole machine assembly and debugging have been completed,and the humancomputer interaction interface has been designed.Based on the developed prototype,the influence of ambient light,exposure time,and conveying speed on the sorting accuracy of shell lotus in the mature stage was carried out.The results show that the ambient light around the field of view has a certain influence on the detection effect,and the conveying speed of the lotus has no effect on the sorting accuracy of the whole machine.When the exposure value is higher than 7000 us,the lotus sorting machine has a high sorting accuracy.The average sorting accuracy is 94.1%.The test results show that the chitosan sorting machine developed in this paper meets the expected design requirements.