Modern Irrigation System Using Convolutional Neural Networks

Abstract

The last twenty years have seen an imbalance between population growth and food production, making agriculture an essential aspect of human survival in the present era. Technological advancements in farming have been crucial to this role. Convolutional Neural Networks (CNN) will be used in this project to identify probable plant illnesses associated with irrigation systems. The objective is to create a model that can correctly categories photos of plants as either healthy or ill and evaluate whether the irrigation system has any bearing on the progression of the disease. In order to do this, a dataset of photos of both healthy and diseased plants that have undergone varied watering conditions will be gathered and pre-processed. Utilizing a deep learning framework like TensorFlow or Flask, the CNN model will be created. Several convolutional layers will be used in the model to extract features from the input photos, and a number of fully connected layers will be used to categorize the images as either healthy or unhealthy. The model will then be trained using the training set by being fed batches of images and their associated labels, and the network's weights will be adjusted depending on the discrepancies between predicted and actual labels. The model will be evaluated on the testing set after training to determine how accurately it makes predictions. Worldwide, crops like tomatoes and paddy are important agricultural products, but because of supply and demand concerns, their prices frequently decline. In order to identify and treat leaf diseases, farmers might not be able to afford agricultural specialists. A low-cost technology that uses image processing can find leaf diseases in tomato and paddy plants to remedy this issue. Farmers can detect infections early and take the necessary action by taking pictures of diseased leaves and comparing them using the CNN algorithm. This technique stabilizes tomato and paddy prices and benefits both farmers and consumers because it is quick, cheap, and practicable throughout the year.