Research A System For Identification And Area Measurement Of Common Cucumber Diseases Based On Artificial Intelligence

In recent years,as China’s rural revitalization strategy has been put into practice,more cucumber plants have been planted.However,as more cucumber plants have been planted,more cucumber diseases have also been discovered.When combined with COVID-19’s impact in recent years,many cucumber plants with disease erosion have suffered losses of varying degrees.With the development of computer technology,it is now possible to use artificial intelligence-based image recognition technology to accurately identify the disease’s type,quickly assess its severity based on the location of the leaf spot,and then use this information to provide growers with detailed treatment recommendations.Treating the disease promptly will improve the yield and quality of cucumbers.In order to give cultivators a new and more practical tool,the research aims to design and implement an artificial intelligence-based system for identifying common cucumber diseases on smartphones through the integration and extension of deep learning models and image processing techniques with We Chat applets.The primary study is described as follows.(1)Creating a collection on cucumber leaf diseases.In order to increase the model’s training accuracy and the dataset’s generalizability,various cucumber growing bases were visited,and various mobile phones were used for picture acquisition and image enhancement processing.to guarantee a higher identification rate of accuracy.(2)The PRe LU function and Dropout regularization were added to the VGG-19 model for the training of diseased leaves,which significantly reduced the overfitting issue in the training process and increased the recognition effectiveness.The VGG-19 model with matching target detection and convolution depth was chosen for training and learning.(3)The severity of the disease is determined by the size of the leaf spot region.The project involves converting the image’s RGB color space to HSV color space,image segmentation to extract the disease from the leaf,multiple traditional image processing operations,identification of the leaf and disease parts using the Canny edge detection model,measurement of the area to obtain the leaf and disease spot’s size,and determination of the disease level using the di The size of the leaf and the disease spot are measured,and the severity of the disease is estimated based on the kind and quantity of the spot.(4)The aforementioned findings were combined,and using We Chat developer tools,an AI-based cucumber common disease identification applet was created.Dosing guidance was provided within the app for various disease types as well as disease levels,making it convenient for cucumber growers.