Design, Analysis, Development of Fruit Diseases Detection and Diagnosis Using Hybrid ML & DL Model

Fruit disorder diagnosis and identification are very vital in keeping agricultural output and best control. This study is a hybrid version of K-Nearest Neighbors and Convolutional Neural Networks collectively to assist within the proper identification and categorization of fruit sicknesses. The CNN features and spatial and hierarchical homes of fruit snap shots are used for function extraction. The extracted characteristics are then labeled the usage of KNN, which is thought for its ease of use and efficiency in working with excessive-dimensional data. To gain an correct evaluation, the dataset contained multiple photographs of end result throughout unique tiers of disorder. Both images taking pictures healthy and dangerous samples of numerous end result are protected within the dataset. By close remark, it is able to be said that the proposed hybrid model achieves first rate accuracy by combining the strengths of each ML and DL techniques; the KNN method has a in a position class potential, whilst the CNN has very strong characteristic mastering skills. When hybrid fashions are in comparison to impartial fashions, they each reap accuracy, recollect, precision, and F1-score differential advanced values. This approach has potential for aiding precision agriculture techniques, decreasing crop losses, and improving early disease detection.