Presenter : Neha

Type: Poster

This paper focuses on the application of machine learning techniques to the geospatial data. Machine learning techniques help to classify any input image based on the features of the image. As an illustration, we have used MNIST dataset which is a dataset of images of Handwritten digits from 0 to 9. The input images are classified using two methods namely, SVM and neural network. The models are trained to learn the features and tested with 10 fold cross validation. SVM resulted in an accuracy of 98% while neural network yielded an accuracy of 99.9%. Given any geospatial image, the machine learning techniques can be used to classify the image after applying image processing techniques such as edge detection filters to the image. This might help in automatically detecting shapes such as junctions and roads in the image.