In this study the disease detections in fruits and vegetables are identified by Machine Learning. The samples of vegetables and fruit leaves with abnormalities are taken into consideration in this inquiry. Farmers can quickly identify illnesses based on the early signs by using these disorder samples of these leaves. Fruit detection and category stays tough because of the texture, color, and form of various fruit categories. To enhance the quality of the vegetable and fruit leaf samples, they are first scaled to 256 by 256 pixels and then subjected to histogram equalization. For the purpose of dividing up dataspace into Polygon cells, the K-means clustering is introduced. Utilizing outline mapping, the edge of leaf sample is retrieved. The relevant characteristics of the leaf samples are extracted using a combination of descriptors, like Grey Level Co-occurrence Matrix. At the end various machine learning techniques viz computer vision have been implemented. Additionally, for faster disease categorization YOLO v3 model is been converted in to open vino IR format this is been noticed that Open-VINO technology raises performance of model by 4 times so plant diseases identification can be executed in short interval of time.