SUMMARY
The process of sorting the quality of papaya fruit is a postharvest problem. So far, humans still do the quality sorting process conventionally or manually. It certainly has weaknesses and limitations, which require a large workforce, and the level of human perception of the quality of papaya varies. Several studies have been carried out regarding the classification of papaya fruit quality, but these studies have accuracy that can still be improved. Therefore, in this study, it is proposed to determine the quality of papaya fruit based on LAB colour and texture features using an Artificial Neural Network (ANN) algorithm. This research consists of six stages: image acquisition, pre-processing, segmentation, morphology, feature extraction and classification. The classification process for the training stage produces the highest level of accuracy in three training scenarios; namely, two techniques have 100% accuracy and 99.58% in the third scenario. Based on the best training scenario selected, the testing process produces 98.88%, the highest accuracy rate with a misclassification error of 1.12% and 69 seconds of computing time. These results indicate that the proposed method can accurately classify papaya quality based on LAB colour and texture features.