Betta Fish Image Identification using Feature Extraction GLCM and K-Nearest Neighbour Classification

Betta fish, known for their striking color patterns and unique fin shapes, present a fascinating domain for image classification. In this study, we propose a comprehensive approach that combines feature extraction using the Gray-Level Co-occurrence Matrix (GLCM) with the K-Nearest Neighbour (K-NN) classification algorithm. 

The project commences with the collection of a diverse dataset of Betta fish images encompassing various species and color variations. Subsequently, GLCM feature extraction is employed to capture texture information from these images, enabling the creation of informative feature vectors. The K-NN classification algorithm is then applied to these feature vectors for species identification. 

This project aims to contribute to the field of computer vision and fish species identification by offering an automated and accurate method for Betta fish classification. The utilization of GLCM for texture analysis and K-NN for classification promises robust performance in distinguishing between different Betta fish species based on their visual attributes. The outcomes of this research could find applications in aquaculture, biodiversity monitoring, and the broader domain of species identification from images.

Keywords—GLCM, K-NN, Betta Fish, Classification, Accuracy