Figure 8 - uploaded by Joonki Paik
Content may be subject to copyright.
Source publication
This paper proposed a new range-based face detection and recognition method using optimized 3D information from stereo images. The proposed method can significantly improve the recognition rate and is robust against object's size, distance, motion, and depth using the PCA algorithm. The proposed method uses the YCbCr color format for fast, accurate...
Context in source publication
Context 1
... B dist , V dist , and A dist , and represent the basic distance, the established value by distance, and the obtained distance, respectively. In order to solve the problem of the low recognition rate due to the uncertainty of size, distance, motion, rotation, and depth, optimized 3D information from stereo images is used. By estimating the position of eyes, the proposed method can estimate the facial size, depth, and pose change, accurately. The result of estimation of facial pose change is shown in Figure 4. Face recognition rate is sensitive to illumination change, pose and expression change, and resolution of image. In order to increase the recognition rate under such conditions, we should consider the pose change as well as the frontal face image. The recognition rate can be increased by the 3D pose information as presented in Figure 5. In order to detect face region and estimate face elements, the multi-layered relative intensity map based on the face characteristics is used, which can provide better result than the method using only color images. The proposed directional blob template can be determined according to the face size. In detail, to fit for the ratio of the horizontal and vertical length of eyes, the template should be defined so that the length of horizontal axis is longer than that of vertical one as shown in Figure 6 (a). The central pixel of a template in a W × H image is defined as P c = ( x c , y c ) . By using W ff × H ff directional template for face components, the average intensity I Dir of 8 -neighborhood pixels is calculated in the central pixel, P c . As a result, the brightness value at P c , I C and the brightness difference value can be obtained. The principal direction, d pr , and its magnitude, d pr , are determined as the direction including the biggest brightness difference as shown in Figure 6 (b). The classified images are trained by PCA algorithm using optimized 3D information component. The block diagram of the proposed optimized PCA algorithm is shown in Figure 8. For the experiment, we extracted 50 to 400 stereo pairs of face images of size 320 ̄ 240. Figure 9 shows the matching result of the left and the right images captured in the distance of 43cm. Composed image shows Figure 9(c) which initializes 20 ̄ 10 block in Figure 9(a), and is searched in the limited region of Figure 9(b). The disparity can be found in the most left and the top regions as shown in Figure 9(c). Facial pose estimation is performed with 9 directional groups at 100cm by using the proposed system as shown in Figure ...
Similar publications
Hexagonal image processing benefits from less computational power and better performance ratios. A lot of research is encompassing this area nowadays. This paper presents the effect of hexagonal image processing in face recognition area where the
proposed hexagonal techniques are applied to the domains of face detection, edge detection and face rec...
Face detection and the numerous applications it leads to are now a part of our everyday lives and can be found in any electronic device. Detecting a face and its facial features in uncontrolled environments, however, come along with two main problems that we address in this paper: handling large variations in pose and facial occlusions. The Active...
This paper describes a robust face recognition algorithm using multiple candidate eye positions to improve recognition. Face recognition systems consist of four major stages. They are face detection, eye detection, face normalisation and face recognition. Most recognition schemes (eg. PCA) assume accurate knowledge of eye positions. By using multip...
Problem statement: Video-based face recognition must be able to overcome the imaging interference such as pose and illumination. Approach: A model is designed to study for face recognition based on video sequence and also test image. In training stage, single frontal image is taken as a input to the recognition system. A new virtual image is genera...
Citations
... Stereo reconstruction methods can be used in this case for additional information estimation. Stereo base face detection methods [6] require additional efforts for depth estimation. It also usually need that cameras have parallel optical axes. ...
This paper presents the algorithm of stereo-faces detection in video sequences. A stereo-face is a face of a man presented by set of images obtained from different points of view. Such data can be used for faces structure estimation. Our algorithm is based on computationally effective method of face detection in mono image. Information about face positions is then combined using sparse stereo matching algorithm. Sparse means that stereo correspondence is estimated not for all scene points but only for points of interest. This allows obtaining the low computation cost of algorithm. We use few criteria to estimate correspondence. These are: epipolar constraint, size correspondence, 3D region of interest constraint and histogram correspondence. Object distance estimation method that does not use projective transformations of stereo planes is also considered.
Glaucoma is the second leading cause of blindness worldwide. Often, the optic nerve head (ONH) glaucomatous damage and ONH changes occur prior to visual field loss and are observable in vivo. Thus, digital image analysis is a promising choice for detecting the onset and/or progression of glaucoma. In this paper, we present a new framework for detecting glaucomatous changes in the ONH of an eye using the method of proper orthogonal decomposition (POD). A baseline topograph subspace was constructed for each eye to describe the structure of the ONH of the eye at a reference/baseline condition using POD. Any glaucomatous changes in the ONH of the eye present during a follow-up exam were estimated by comparing the follow-up ONH topography with its baseline topograph subspace representation. Image correspondence measures of L1-norm and L2 -norm, correlation, and image Euclidean distance (IMED) were used to quantify the ONH changes. An ONH topographic library built from the Louisiana State University Experimental Glaucoma study was used to evaluate the performance of the proposed method. The area under the receiver operating characteristic curves (AUCs) was used to compare the diagnostic performance of the POD-induced parameters with the parameters of the topographic change analysis (TCA) method. The IMED and L2-norm parameters in the POD framework provided the highest AUC of 0.94 at 10 degrees field of imaging and 0.91 at 15 degrees field of imaging compared to the TCA parameters with an AUC of 0.86 and 0.88, respectively. The proposed POD framework captures the instrument measurement variability and inherent structure variability and shows promise for improving our ability to detect glaucomatous change over time in glaucoma management.
