Towards a Real-Time Oil Palm Fruit Maturity System Using Supervised Classifiers Based on Feature Analysis

Remote sensing sensors-based image processing techniques have been widely applied in non-destructive quality inspection systems of agricultural crops. Image and analysis were performed with computer vision external grading by general standard steps, such as acquisition, pre-processing segmentation, extraction classification characteristics. This paper describes the design implementation a real-time fresh fruit bunch (FFB) maturity system for palm oil based on unrestricted remote (CCD camera sensor) using five multivariate (statistics, histograms, Gabor wavelets, GLCM BGLAM) to extract characteristics incorporate information species FFB testing. To optimize proposed solution terms performance reporting time, supervised classifiers, support vector machine (SVM), K-nearest neighbor (KNN) artificial neural network (ANN), evaluated via ROC AUC measurements. The experimental results showed that maturation real time provided significant result. Although SVM classifier is generally robust classifier, ANN has better due natural noise data. highest precision was obtained basis BGLAM algorithms texture fruit. In particular, algorithm feature technology largely high test accuracy over 93% an image-processing 0,44 (s) detection species.