Diseases in plants pose a major impact on the crop yield. They severely affect the quality and quantity of agricultural crops. Therefore accurate detection of infection in plants in a timely manner is important to limit the transmission of the disease and to enhance the crop productivity. Manual examination of the plant diseases requires a lot of time, effort and cost can even lead to faulty treatments. In order to counter this problem, many methods based on image processing and machine learning methods have been suggested. This paper implements a deep learning method based on convolutional neural networks(CNN) combined with long short-term memory(LSTM) network for identifying diseases in plants. It makes use of the PlantVillage dataset which consists of images of leaves of healthy and diseased plant crops belonging to 14 crop species. The proposed model achieves an accuracy of 95.11%, which suggests that CNN model used along with LSTM network for classification can help to enhance the accuracy of the CNN model. The proposed system can thus help the farmers to detect plant diseases easily.