The Wheat head detector and health classifier that is proposed in this work is a system that takes an aerial image of field wheat, at a certain distance, as an input and detects all the wheat-heads in it. After detecting them, we were able to predict the health of the crop, i.e. healthy or diseased, identifying diseased wheat heads by processing the input images. This work is based on machine learning and uses state-of-the-art machine learning algorithms i.e. Residual Neural Network (ResNet-50) and a convolutional neural network (CNN) model. Along with this, two datasets were used to train these models. Global wheat head detection (GWHD) dataset was used to train the ResNet-50 model for detecting wheat-heads and a self-constructed dataset is used to train the wheat health classifier (CNN).