knn_image_classification

KNN classification algorithm is one of the well known classification methods in data engineering and image processing. This algorithm is basically a start point to use ML algorithms in the world of image processing. In this project, the KNN algorithm has been used to classify 3 kinds of animals namely, cat, dog, and panda. The animal dataset including 3000 images(1000 images per animal) has been used. The KNN algorithm has been written in python3. The results show that the KNN performance in image classification can be improved by increasing the number of nearest neighbors K to a specific number. Having said the performance cannot be improved by increasing K beyond the extracted criteria. Having said that the performance has a direct relation with the data quality.