Development Of Apricot Mushroom Appearance Quality Grading And Sorting Device Based On Machine Vision

The fleshy and nutritious apricot mushroom has high edible,nutritional and economic values,and is one of the economic edible mushroom species cultivated in factory in China,with great market potential.At present,the industry has been grading and sorting by hand,with high work intensity and low grading and sorting efficiency,and the quality fluctuates with the subjective ideas of workers,which cannot achieve the effect of scale,uniformity and high efficiency of sorting.This topic applies the machine vision intelligent grading and sorting technology to the apricot mushroom sorting industry for the first time,and proposes a machine vision-based apricot mushroom appearance quality grading and sorting device,which can effectively solve the problems of low efficiency of manual sorting,quality fluctuation and increase of labor cost,and is of great significance to the application and development of apricot mushroom industry.The research content of this topic is as follows.(1)Establish the grading criteria based on the apricot mushroom in combination with the relevant grading standards of the country and enterprises,and complete the experimental research on physical characteristics.Determine the grading standards for each size of Abalone mushroom based on 6 characteristic elements such as length,diameter,curvature,uniformity,cap defect and mushroom body color,complete the physical characteristics test related to Abalone mushroom,establish the ultimate shape size model of Abalone mushroom using Solid Works 2018,and determine the physical characteristics such as quality,shape size,friction coefficient,safe fall height and density of Abalone mushroom.(2)Complete the design of mechanical system of apricot mushroom grading and sorting device based on the principle of differential separation.Introduce and model the principle of differential separation of Abalone mushroom,simulate and analyze the differential separation model with ADAMS 2018,complete the design and development of the mechanical system virtual prototype with Solid Works 2018,and build the test prototype to complete the principle verification test.(3)Complete the construction of the apricot mushroom image acquisition system and image processing algorithm research.Select key components of the image acquisition system and complete camera calibration and correction.Research on the algorithms and principles of image correction transformation,image enhancement,edge detection and color space conversion.Through the indexes of code operation time,peak signal-to-noise ratio and edge line completeness,the best processing effect algorithm is determined,and an improved Canny operator with bilateral filter instead of Gaussian filter as image smoothing filter and Ostu maximum inter-class variance method instead of fixed double threshold segmentation is proposed for the shortage of traditional Canny operator.Finally,a median filter is selected for image smoothing,an improved Canny operator is used as an edge detection algorithm to extract the length,diameter,curvature and uniformity with a minimum external rectangle,and the apricot mushroom RGB image is converted to HSV color space to obtain H,S and V components to extract the mushroom cap defect and mushroom body color.(4)Complete the vision software interface development using HALCON in conjunction with C# in.NET platform,including 5 modules function development such as image acquisition,image processing,serial communication,product configuration and database management.The software and the working performance of the device were tested online,and the results showed that the grading accuracy of the diameter of Abalone mushroom was 83%,and the remaining characteristic elements could reach more than 85%,and the overall grading accuracy of the device for each specification of Abalone mushroom was more than 90%.