Carrot Quality Classification Based on Color and Texture Features Using Artificial Neural Network Method

Carrots are popular vegetable plants that are usually consumed by the public. Determination of quality using the visual of human eye is considered to have many shortcomings. In previous studies, the carrot classification process had been carried out using a certain method. However, the level of accuracy resulting from several previous studies is still lacking because the processes and methods used are considered to be inaccurate, so innovation is needed by using processes and methods that are more precise to obtain classification results with a better level of accuracy. Therefore, this research proposes a classification of carrot quality based on color and texture features using an artificial neural network method. The proposed method consists of 6 stages, namely image acquisition, preprocessing, segmentation, morphological operations, feature extraction, and classification using artificial neural networks. In this study, quality is divided into three classes, namely feasible, less feasible, and not feasible using 300 carrot image datasets. The results obtained in the testing process obtained an accuracy of 100%, a misclassification error of 0%, and a computation time of up to 55 seconds. Based on the test results it can be seen that the proposed method can classify the quality of carrots accurately.