Plant leaf disease identification is a major challenge in the agricultural sector but their rapid identification is difficult because of lack of necessary infrastructure. Faster and accurate prediction of leaf diseases in plants could help in early treatment and reduce economical loss to the farmers. Modern advanced developments in deep learning have increased the performance and accuracy of object detection and recognition systems. Disease detection in plant leaves detects the infected part of the leaf and predicts the disease present in the leaf and also displays the remedy that can be used to cure and provide proper care to the plants. In this product, Convolutional Neural Network (CNN) models is used to perform plant disease detection and diagnosis using simple leaves of healthy and infected plants through deep learning. CNN is a deep neural network originally designed for image analysis. CNN always contains two basic operations, namely convolution and pooling. The convolution operation using multiple filters is able to extract features (feature map) from the data set, through which their corresponding spatial information can be preserved. The pooling operation, also called subsampling, is used to reduce the dimensionality of feature maps from the convolution operation. Training of the models was performed with the use of an open dataset of more than 54,000 images of the plants of 14 crop species. The trained model achieves an accuracy of 68.8% on a held-out test set. The proposed product can effectively identify different plant leaf diseases and can be used as an advisory or early warning tool for better cultivation of the plants.