Application Research on Detection of Edge Characteristics of Tea Garden Disease Based on UAV Inspection

To achieve reliable detection of diseases in tea gardens, a detection approach based on unmanned aerial vehicle (UAV) inspection was put forward. UAV aerial imaging technology is employed to collect image information of tea garden diseases, and machine learning methods and deep learning network models are utilized to identify and classify diseases of tea garden crops. The feature-based approach mainly extracts key points from regions with distinct features such as structure and texture in both the reference image and the image to be registered, and constructs texture, color, and other parameters that conform to the characteristics of disease and pest information based on the neighborhood information of the key points. Based on the method of edge feature detection, a distributed reconstruction model of the edge of tea garden diseases is established. According to the UAV inspection results combined with image analysis methods such as wavelet analysis, gray analysis, and histogram analysis, the shape, color, and texture feature points of tea garden diseases were classified and recognized, thereby achieving effective identification of diseases and pests in tea gardens. The simulation results indicate that the method can accurately identify and locate tea pests and diseases with high similarity across some diseases, different seasons, and different regions, particularly the expression of color tea image features.