Lip segmentation based on Lambertian shadings and morphological operators for hyper-spectral images

摘要: Abstract Lip segmentation is a non-trivial task because the colour difference between lip and skin regions maybe not so noticeable sometimes. We propose an automatic technique for hyper-spectral images from imaging prototype with medical applications. Contrarily to many other existing methods, we do use space transformations localise area. As input image, first time parametric blood concentration map computed by using narrow spectral bands. Our method mainly consists of three phases: (i) each subject generate subset face enhanced different simulated Lambertian illuminations, then (ii) perform on image constrained morphological operations, finally (iii) extract features Fourier-based modeled boundaries selecting candidate. Experiments testing our approach are performed under controlled conditions volunteers public dataset. Results show effectiveness algorithm against low range, moustache, noise.