Project: Mars Surface Image Classification using Machine Learning Algorithm

The project's goal was to use machine learning techniques to categorize the terrain on Mars using information from Mars missions such as the Mars Reconnaissance Orbiter and Mars Science Laboratory. The Project identifies the challenges of classifying images due to the lack of helpful metadata, as the metadata is mostly related to the features of the images rather than providing useful classification information. The project aims to use machine learning algorithms such as Decision Trees, Linear Regression, and K-nearest neighbors (KNN) to classify the images and predicts the type of terrain and further compares the accuracy of each algorithm. Also provides an explanation of the dataset used, which consists of captioned and non-captioned images with a JSON file containing details such as acquisition date, local Mars time, latitude, longitude, etc. Finally, the project's major objective is to provide a system that astronauts can use to classify various terrains in new photographs.