Precision agriculture demands the recognition of plant species with help of distinct leaf characteristics such as form, texture, border, and colour etc. Plant leaf recognition has drawn special interest in the domain of computer vision.Existing models classifyspecific types of plants. These models classify limited number of plants at time. But there is need to recognize maximum number of plants in single model. This paper deals with the development of the recognition model using transfer learning techniques, which classify the plants into respective category. The implemented model classifies 21 different plant types using MobileNetV2, a transfer learning approach and achieved 98% accuracy.

Keywords: CNN, MobileNetV2,Plant leaf classification, Transfer Learning