Adopting level set theory based algorithms to segment human ear
This work addresses the segmentation challenge in ear biometrics, which is incremental as it applies existing PDE-based methods to a specific domain.
The paper tackled the problem of segmenting human ear images for biometric identification by applying level set theory-based algorithms, achieving a comparison of their efficiency on test images.
Human identification has always been a topic that interested researchers around the world. Biometric methods are found to be more effective and much easier for the users than the traditional identification methods like keys, smart cards and passwords. Unlike with the traditional methods, with biometric methods the data acquisition is most of the times passive, which means the users do not take active part in data acquisition. Data acquisition can be performed using cameras, scanners or sensors. Human physiological biometrics such as face, eye and ear are good candidates for uniquely identifying an individual. However, human ear scores over face and eye because of certain advantages it has over face. The most challenging phase in human identification based on ear biometric is the segmentation of the ear image from the captured image which may contain many unwanted details. In this work, PDE based image processing techniques are used to segment out the ear image. Level Set Theory based image processing is employed to obtain the contour of the ear image. A few Level set algorithms are compared for their efficiency in segmenting test ear images.