Identifikasi Jenis-Jenis Burung Lovebird Menggunakan Pengolahan Citra Digital Dengan Metode K-Means Clustering

Abstract

Identification process type of bird lovebird, usually determined by parameters characteristic feather color of birds lovebird. Parameters to define the type of bird lovebird usually done manually by using the sense of sight. This way has a weakness that is, it takes a long time and results in an invalid identification. Based on that weakness, the researcher designed an identification system type of lovebird birds by using the Matlab programming language. Identify the type of lovebird bird conducted by researchers, namely color analysis of lovebird bird feathers. Image data were taken in the form of bird samples, lovebird josan, albino, and lutino by using a DSLR camera and identification algorithm K-means Clustering. The system is divided into two stages, namely the training stage and the testing phase. The training stage uses 30 bird images recognized according to type resulting in an accuracy level of 100%. While The testing phase uses 24 bird images, 22 images of lovebird birds identified according to their type, and two recognizable lovebirds images but not according to its kind resulting in an accuracy level of 91,67%.