Research on Field Cotton Yield Prediction System Based on Improved Yolov5 Cotton Detection Model and Android Development

Cotton is an important cash crop in China, which has important applications in agricultural and industrial production. The prediction of cotton yield helps in economic regulation and adjustment of planting patterns, improving production returns, while traditional manual yield measurement methods require a significant amount of time and labor costs. To address this issue, this study selected cotton images after spraying defoliating agents as the research object and constructed relevant datasets. At the same time, based on the calculation formulas for the number of cotton plants, cotton bolls, and single boll seed cotton weight per unit area, an improved YOLOv5 algorithm model was used as the core algorithm to design a cotton yield prediction system based on Android mobile devices. Obtain image information by selecting mobile phones to take photos or selecting to call photo albums, and carry out data analysis and processing on the target image to achieve cotton yield prediction. The cotton bolls in the image are detected by the cotton detection box, and the cotton yield per hectare is automatically calculated according to different soil type. Compared with the actual yield, the average error between the actual yield and the predicted yield of seed cotton and lint cotton is 122.01 and 57.98kg/hm2, The average yield error per mu is 8.134 and 3.865kg, and the model has higher accuracy on the mobile phone. Compared with the original YOLOv5 model, the accuracy P (%) and recall R (%) have increased by 19.58 and 16.84, respectively, with values of 90.95 and 73.16%. After comparison and testing on three types of mobile phones, the system runs smoothly and the yield prediction results are not significantly different. Research has shown that the developed cotton yield prediction system has good performance in field yield measurement results and algorithm operation, providing new methods and ideas for cotton yield prediction, and has certain application value. At the same time, it provides reference methods for field yield prediction of other crops.