About this Project

Objective: Create a highly accurate model that can predict whether an image is a cat or a dog using Digital Signal Processing (DSP) techniques, image processing, and deep learning.

DSP Problem of Focus: We have trained and compared various models such as k-nearest-neighbors(KNN), decision tree, and convolution neural networks(CNNs) to find the most accurate and robust model for image filtering and classification.

Bigger Picture: Whether it's creating a database of animals to be able to distinguish from, analyzing medical images, or designing autonomous systems, image classification using machine learning is a very important topic in modern society. If we are able to create a model that can classify two animals: cats and dogs, we can expand upon it and use similar techniques in more complex systems. For example, we can expand upon the DSP techniques and CNNs from this project to classify types of cats and dogs, emotion based on facial expressions in pictures, or distinguish between pedestrians/hazardous objects in the road scene for autonomous vehicles. Classifying cats versus dogs is like touching the tip of the iceberg in the world of image processing. It is an abstraction of larger world problems, but complex enough to build a foundation in image processing and deep learning using DSP.