The agricultural sector is vital to human existence. Approximately sixty percent of people work in agriculture, either directly or indirectly. Farmers are afraid of increasing their agricultural productivity day by day because the existing system lacks the technologies needed to identify diseases in different crops in an agricultural environment. Early detection is crucial because crop diseases have an impact on the growth of the corresponding species. One of the more significant agricultural commodities in the world, tomatoes are used in many different cuisines all over the world. For the purpose of detecting leaf disease, numerous conventional models have been presented. Lately, nevertheless, both supervised and unsupervised learning tasks have demonstrated improved results. Accordingly, a convolutional neural network (CNN)-based work had been put forwarded here. The dataset, which includes six distinct disease kinds that affect tomato leaves and has a sample size of 6000, was obtained from Kaggle.com and used in the experiment. The dataset’s leaf photos are fed into the leaf disease detection system, where they undergo pre-processing before features are extracted. The final component of CNN, the softmax classifier, receives the extracted features. With this dataset, an accuracy of 99.3% was obtained overall. We have utilized Flask for the entire framework and testing, CNN for training and validation, Python for model building, and the OpenCV package for image capture and storage. Glaucoma is a condition that affects the optic nerve and can be linked to diabetes, which raises pressure inside the eye. If not caught early, it can lead to significant loss. Machine learning can automate the detection of this disease. With the plant leaf disease detection, this paper also explores how CNN is used in mobile apps to effectively detect glaucoma, focusing on identifying the important factors for better and accurate detection systems.