Identifikasi Citra Luka Abalon (Haliotis Asinina) Menggunakan Gray Level Co-Occurrence Matrix dan Klasifikasi Probabilistic Neural Network

Abstract

Wounded abalone causes a high mortality rate in abalone life and big loss for breeder. Nowadays, objects can be performed by processing digital images. In this research, Gray Level Co-occurrence Matrix (GLCM) is used to identify wounded abalone. Identification is conducted using the five elements of grayscale image: energy, homogeneity, contrast, correlation and entropy. The identification result using Probabilistic Neural Network (PNN) produce average accuracies of 85% using good images and 37% using bad images.