Intelligent classification of Common Carp (Cyprinus carpio) based on freshness using the combined of image processing techniques and adaptive neuro-fuzzy inference system

This paper proposes an image processing method in combination with the intelligent adaptive neuro-fuzzy inference system (ANFIS) for classifying common carp bodies based on the freshness factor during the storage period under ice-covered conditions. In doing so, after image acquisition, for pre-processing, the images were transferred to various color channels and the statistical properties of the image texture were extracted. In order to increase the speed and accuracy of classification, the principal component analysis method (PCA) was used to reduce the dimensions of the features. Evaluation of the classifier was performed to identify the freshness level using statistical indices such as accuracy, precision, sensitivity, specificity and area under the curve (AUC). The values of these indices for classification using ANFIS for the test data were obtained as 90.33, 79.1, 77.36, 92.57 and 84.97, respectively. The acceptable results obtained from fish images showed that the current method has the ability for quick online detection of fish freshness in the food industry as a low-cost, simple and non-destructive method.