Tomato plant is very prone to suffer the attack of the foliar diseases that are able to affect adversely the yield and the quality of the produce. Nevertheless, the sustainable agricultural practices depend on early identification and correct categorization of these diseases. Furthermore, these five deep learning architectures were compared against one another to be able to detect and classify diseases that have affected the tomatoes such as CNN, VGG16, ResNet50, InceptionV3 and MobileNet. To compare and contrast every model with the required data pre-processing and augmentation, this was done using a large dataset. And ResNet50 and VGG16 provide such accuracy and can take varying diseases of rans (viral, bacterial and fungal). The proposed system is a real time mobile plant disease diagnosis and monitoring system that has the flexibility of scalability, crop loss preventing capabilities, reduction of use of chemicals as well as support to precision agriculture.