A Survey on Optimization based Ear Recognition system
Payal Verma
, Prof. Sandeep Patil
Biometric are automated methods of recognizing a person based on a physiological or behavioral characteristic. Most of the well known ear biometric techniques have focused on recognition on manually cropped ears and have not used automatic ear detection and segmentation. This is due to the fact that detection of ears from an arbitrary profile face image is a challenging problem as ear images may vary in scale and pose (due to in-plane and out-of-plane rotations) under various viewing conditions. However, for an efficient ear recognition system, it is desired to detect the ear from the profile face image in an automatic manner. There exist few techniques in the literature which can be used to detect ear auto-matically. A detailed review of these techniques is as follows. The first well known technique for ear detection is due to Burge and Burger [1]. It has detected ears with the help of deformable contours. But contour initialization in this technique needs user interaction. As a result, ear localization is not fully automatic. Hurley et al. [2] have used force field technique to get the ear location. The technique claims that it does not require exact ear localization for ear recognition. However, it is only applicable when a small background is present in ear image. In [3], Yan and Bowyer have used manual technique based on two-line landmark to detect ear where one line is taken along the border between the ear and the face while other line is considered from the top of the ear to the bottom. The 2D ear localization technique proposed by Alvarez et al. [4] uses ovoid and active contour (snake) [5] models. Ear boundary is estimated by fitting the contour of an ear in the image by combining snake and ovoid models. This technique requires an initial approximated ear contour to execute and hence cannot be used in fully automated ear recognition system. There is no empirical evaluation of the technique.
"A Survey on Optimization based Ear Recognition system", IJSDR - International Journal of Scientific Development and Research (www.IJSDR.org), ISSN:2455-2631, Vol.2, Issue 6, page no.436 - 443, June-2017, Available :https://ijsdr.org/papers/IJSDR1706066.pdf
Volume 2
Issue 6,
June-2017
Pages : 436 - 443
Paper Reg. ID: IJSDR_170357
Published Paper Id: IJSDR1706066
Downloads: 000347173
Research Area: Engineering
Country: Durg, NA, India
ISSN: 2455-2631 | IMPACT FACTOR: 9.15 Calculated By Google Scholar | ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 9.15 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Publisher: IJSDR(IJ Publication) Janvi Wave
