SKIN CANCER IMAGES CLASSIFICATION USING NAÏVE BAYES

Abstract

It might be well known that skin cancer is one of the foremost perilous sorts of cancer, essentially, there are two sorts of skin cancer named malignance and non-malignance. In current paper the system is designed for the exposure and classification of skin cancer with high precision and exactness by using Naïve Bayes which capable to diagnoses different sorts of cancer in a human skin. It objects at utilizing more significant information to develop the skin cancer and assistance physicians in the clinical to diagnosis and exact detection of disease. The designed system has an important role in to avoid errors during the while identification and sorting of cancer. This is encouraged by probable execution enhancement in the overall automatic and giving confidence in decision-making and hasty detection of skin cancer. This technology is of extraordinary and financial significance to doctors. The system is separated into two sorts (system with pre-processing and another system without pre-processing). The former system contains the following stages: image acquisition, preprocessing, and classification, whereas the last system contains of image gaining and classification. The results from this paper show that the model using (NB) without any pretreatment has an average level of accuracy 70.15%, however with preprocessing has an accuracy of 69.69%. The diminished precision returns to the reason that the skin pictures that are taken to the skin are as well near and they don't require any handling.