Detecting pests and diseases in crops has always been difficult for farmers. They can significantly reduce crop yield if not detected at an early stage. Pests and diseases occur in different phases of crop development. As a result, a continuous monitoring system is essential. Various advancements in the field of deep learning and machine learning have aided in the monitoring and diagnosis of crop- related diseases, resulting in improved production. This research aims to work in the area of disease and pest recognition using machine learning and deep learning. This paper examines and summarizes different techniques that researchers used in their studies on disease and insect detection for the health monitoring of crops. The research paper puts forward two methodologies: First to identify diseases in cotton leaves with the help of three image classification models namely VGG16, Resnet50, and Inception V3, and attained an accuracy of 99.58%, 85.03%, and 95.38% respectively on a custom dataset of 875 images. Second to detect pests in crops using the VGG16 and Inception V3 and achieved an accuracy of 99.78% and 97.96% respectively on the pest dataset available on Kaggle.