Ear Feature Extraction Methods - A Review
Lilis Yuningsih, Gede Angga Pradipta, Dandy Pramana Hostiadi, Roy Rudolf Huizen, Putu Desiana Wulaning Ayu · 2022 4th International Conference on Cybernetics and Intelligent System (ICORIS) · 2022
Research on system authentication with ears as the object have been widely studied and developed. The object of ears has some excellences such as being unchangeable in structure due to the age, unique, and easy to be obtained. The feature extraction process plays the most important role that highly determines the accuracy level from the recognition and classification of image. The research on ear recognition has been continuously developed to reach its completeness. This paper presents a review on research that has been done and divided into three categories: 3D feature extraction, appearance-based feature extraction, geometric feature extraction, and hybrid feature extraction. This paper also presents any advantages and disadvantages of each category as well as a number of challenges frequently faced by previous research related to the feature extraction.