Navel Orange Blemish Identification for Quality Grading System

A novel automated blemish detection system for ripe and unripe oranges is proposed in this paper. The algorithm is unique in that it does not rely on the global variations between pixels depicting the colours of an orange. By utilizing a priori knowledge of the properties of rounded convex objects, we introduce a set of colour classes that effectively 'peels-off' the orange skin in order of increasing intensity layers. These layers are then examined independently, allowing us to scrutinize the skin more accurately for any blemishes present locally at the layer's intensity variation range. The efficacy of the algorithm is demonstrated using 170 images captured with a commercial fruit sorting machine as the benchmarking test set. Our results show that the system correctly classified 96% of good oranges and 97% of blemished oranges. The proposed system does not require any training.