Design Of Cucumber Leaf Moisture Content Detection System Based On Image Processing Technology

Cucumber is one of the main crops planted by farmers in agriculture,with a wide distribution area,but it is still in the initial stage of research on the non-destructive detection of cucumber leaf moisture content.Based on this situation,this paper uses image processing techniques to carry out detection,respectively studied the immune genetic+OTSU algorithm and the standard PSO+OTSU image segmentation algorithm,and improved the standard PSO+OTSU image segmentation algorithm,and proposed a PSO+OTSU image segmentation algorithm based on image histogram,using the histogram of the blade image improves the initial position of the particle swarm of the PSO algorithm and the update formula of the inertia weight coefficient,which improves the efficiency and the accuracy of image segmentation.The multivariate linear regression algorithm,BP and RBF neural networks were used to mathematically model the moisture content of cucumber leaves,R2 and RMSE were used to evaluate the model,and the optimal model was selected to predict the moisture content of cucumber leaves.This article completed the following design and research tasks:(1)Perform image preprocessing operations on cucumber leaf images.Image preprocessing is mainly for image graying,filtering,segmentation,morphological operations and image restoration.This paper improves the standard PSO+OTSU image segmentation algorithm,proposes an improved PSO+OTSU algorithm,and conducts a large number of experiments to prove that the improved PSO+OTSU algorithm can improve the accuracy of image segmentation,reduce the number of algorithm iterations,and improve the image Segmentation efficiency.(2)Extract color,shape,and texture characteristic parameters related to the moisture content of cucumber leaves,and use Pearson correlation coefficients to select the characteristic parameters with larger correlation coefficients.(3)This article uses three different methods to establish the model of cucumber leaf moisture content,and uses the determination coefficient R2 and root mean square error RMSE to evaluate the model,selects the model with the highest degree of fit,and uses it as the prediction of cucumber leaf model.(4)This article uses MATLAB to design a detection system,which respectively implements five functional modules:image import,image preprocessing,feature value extraction,model prediction,and result analysis,and implements the cucumber moisture content detection function based on the PC platform.In this paper,the detection method of cucumber leaf water content based on image processing technology is designed.The cucumber leaf water content is predicted based on the cucumber leaf image,and the PSO+OTSU image segmentation algorithm is improved.The histogram is used to improve the initial position of the PSO particle swarm,and improves the gray scale model in the OTSU algorithm,the improved algorithm increases the accuracy of image segmentation,reduces the number of image segmentation iterations,improves the efficiency of image segmentation.Models the cucumber leaf moisture content based on the extracted feature values,and using MATLAB to complete the design of the cucumber water content detection system based on the PC platform.