Abstract <p>Dental caries lead to extreme discomfort, mutilation and even casualty in a few cases. The severity of dental caries depends on their depth in the tooth i.e., the extent to which they have damaged the tooth. This leads to their classification into four classes namely, incipient, moderate, advanced and severe. Identifying the caries severity is a matter of perception and to do so in an early stage is important to cure them completely. In this paper, we propose an intelligent fuzzy inference system to classify dental caries into the above-mentioned severity types. The dental X-rays of infected tooth are segmented using color image segmentation based on RGB color model and then four attributes R-G, R-B, G, and B are selected after careful observation of ground truth images. The fuzzy rules are designed based on the feature vector comprising of these attributes and fed as an input to the fuzzy inference system. The output of the fuzzy inference system is the required classification based on the severity. The dataset consisted of 1000 periapical dental images, out of which 700 images were used for training and 300 images were used for testing. The accuracy of the proposed severity classification method is 95%. The better performance metrics of the proposed fuzzy approach as compared to current state of art classifiers ascertain its efficacy in classifying dental X-ray images of caries infected teeth, into different severity types.</p>