A Review on Preprocessing, Segmentation and Classification Techniques for Detection of Skin Cancer

Abstract

About 2 to 3 million new cases of skin cancer are identified every year, making it one of the most common forms of cancer. Skin lesions are the result of anomalous cell reproduction on the skin, and examining them manually is a tough, instinctive, and arduous task. Computer-Aided Diagnosis (CAD) technologies allow practitioners to become more proficient in their investigative abilities while at the same time reducing the amount of time necessary to carry out an appropriate diagnosis. However, the restricted accessibility of several advanced computer-aided diagnoses (CAD) systems that are easier to use has resulted in a significant amount of uncertainty over the rapid, non-invasive, and reliable detection of diseases. According to the existing literature, not nearly enough effort has been put into creating CAD systems for skin lesions. With a primary focus on pre-processing, segmentation, and classification of skin lesions in dermoscopic images, the goal of this article is to give a complete and critical bibliography addressing the published literature on this topic. The purpose of this paper is to provide detailed yet necessary references. The main goal is to give researchers who are new to this field and don't know enough about it to understand all of its technical details an overview. In addition, it identifies some of the research challenges that can be identified in the literature and need to be addressed by future researchers to investigate innovative pathways in this area.