Fungus Detection and Identification using Computer Vision Techniques and Convolution Neural Networks

In this paper we have presented a fungus detection system. As we all know that Fungus is a very big risk for the human Life as it can come in contact to humans through food, water, Air etc. Exposure to high concentrations of fungal spores from air can cause many health problems, such as bronchial asthma and other deadly diseases. Fungus is the main risk for food logistics. We aim to build a model which can correctly detect and identity fungus. Datasets plays a critical role throughout the procedure. For detection purposes, an accurate dataset is required. The dataset we are collecting was made using an optical acquisition system. The main task of our project is the image analysis. Image analysis can be done using computer vision techniques or with the help of machine learning. We have explained both the methods of analysis of CVT using Histogram oriented gradients (HOG) and Support Vector Machine and Convolution neural networks (CNN) in detail with various processes and layers involved.