Difference in Lights and Color Background Differentiates the Color Skin Model in Face Detection for Security Surveillanc
Face detection with variable lights and color background makes it more difficult to detect the originality of the person in the image. Subject does not look directly into the camera; when the face is not held in the same angle, the system might not recogn
- PDF / 229,594 Bytes
- 9 Pages / 439.37 x 666.142 pts Page_size
- 64 Downloads / 157 Views
Abstract Face detection with variable lights and color background makes it more difficult to detect the originality of the person in the image. Subject does not look directly into the camera; when the face is not held in the same angle, the system might not recognize the face. In this paper, we are considering various live studies where security surveillance ought to be a first preference of our own lives. Few studies have taken as source input study which helped us for better outcome. Further algorithm designed to get significant result is least expected to perform well on small sample data. Keywords AdaBoosT training algorithm model Color Similarity Image (CSI)
⋅
⋅
Skin color segmentation
⋅
Skin color
1 Introduction The rising number of face recognition applications in everyday life where image segmentation and video-based recognition methods are becoming very important research area. Generally, effects of pose, illumination, facial expression, and occlusion are such issues mostly studied in face recognition. So far, very little has been done to investigate the effects of compression on face recognition, even though the images are mainly stored and then translated into a compressed format. Still pictures have been experimented so often, but only in uncompressed image formats, whereas in videos, mostly research deals with basic issues of tracking and recognizing faces where still uncompressed images have taken as library and compressed video as probes. D. Chawla (✉) Pacific Academy of Higher Education & Research University, Udaipur, India e-mail: [email protected] M.C. Trivedi ABES Engineering College, Ghaziabad, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2018 G.M. Perez et al. (eds.), Networking Communication and Data Knowledge Engineering, Lecture Notes on Data Engineering and Communications Technologies 4, https://doi.org/10.1007/978-981-10-4600-1_12
127
128
D. Chawla and M.C. Trivedi
In this paper, we have focused on change in lights and background base which actually makes image resolution different in recognizing the effect on each individual faces. Also, the change can reflect to the change with aging, change respect to plastic/cosmetic surgery, or with any other formation of difficulty level to recognize the face. We have tried to demonstrate the system through proposing an algorithm in order to recognize each individual which is explained and its related work in Sect. 3, the experimental result and comparison with other color model and in Sect. 4 followed by the conclusion and future scope.
2 Literature Review As explained in (M. Singh 2014), with the change in time and age, circumstance changes reflect on each individual faces, skeleton structure, muscle mass, and body fat [1]. Image-based techniques in (Philippe Carré 2014) have formulated color alterations with algebraic operations [2]. The generalized linear filtering algorithms defined with quaternions and define a new color edge detector. The group of authors in (V.V. Starovoitov 2002) have trained a process to
Data Loading...
