We present here an application of standard biometric recognition techniques in the context of ear feature extraction. The objective is to apply image processing algorithms that detect and extract general descriptors for ear images in order to recover the Pinna Related Transfer Function that is most appropriate for pinnas captured in an input query image. In particular, this project includes a prototype implementation of the input query, ear detection, and nearest neighbor matching process via MATLAB.
The pinna (outer ear) plays an important role in localizing the elevation position of sound sources as it generates a series of elevation cues while filtering the acoustic signal. This can be described via a frequency response function called the head related transfer function (HRTF). Different individuals have distinctive HRTFs since the biometric parameters vary significantly in relation to size, shape, and orientation.