Relevance weighted (2D)2LDA image projection technique for face recognition application

2009 ◽  
Author(s):  
W. Sanayha ◽  
Y. Rangsanseri
Keyword(s):  
Face Recognition ◽  
Projection Technique ◽  
Image Projection

scholarly journals Relevance-Weighted (2D)2LDA Image Projection Technique for Face Recognition

ETRI Journal ◽  
2009 ◽  
Vol 31 (4) ◽  
pp. 438-447 ◽  
Author(s):  
Waiyawut Sanayha ◽  
Yuttapong Rangsanseri
Keyword(s):  
Face Recognition ◽  
Projection Technique ◽  
Image Projection

Self-Adapting Patch Strategies for Face Recognition

2019 ◽  
Vol 34 (02) ◽  
pp. 2056002 ◽  
Author(s):  
Zhi-Ming Li ◽  
Wen-Juan Li ◽  
Jun Wang
Keyword(s):  
Face Recognition ◽  
Local Binary Pattern ◽  
Greedy Algorithms ◽  
Discrete Wavelet ◽  
Projection Technique ◽  
Sparse Constraint ◽  
Integral Projection ◽  
New Methods ◽  
Information Combining ◽  
Local Binary Pattern Feature

In this paper, we propose two self-adapting patch strategies, which are obtained by employing the integral projection technique on images’ edge images, while the edge images are recovered by the two-dimensional discrete wavelet transform. The patch strategies are equipped with the advantage of considering the single image’s unique properties and maintaining the integrity of some particular local information. Combining the self-adapting patch strategies with local binary pattern feature extraction and the classifier of the forward and backward greedy algorithms under strong sparse constraint, we propose two new face recognition methods. Experiments are run on the Georgia Tech, LFW and AR face databases. The obtained numerical results show that the new methods outperform some related patch-based methods to a larger extent.

Augmented reality–guided neurosurgery: accuracy and intraoperative application of an image projection technique

2015 ◽  
Vol 123 (1) ◽  
pp. 206-211 ◽  
Author(s):  
Leila Besharati Tabrizi ◽  
Mehran Mahvash
Keyword(s):  
Augmented Reality ◽  
Neurosurgical Procedures ◽  
Fiducial Markers ◽  
Projection Technique ◽  
Image Guided ◽  
Augmented Reality System ◽  
Brain Surface ◽  
Image Projection ◽  
The Mean ◽  
Projection Error

OBJECT An augmented reality system has been developed for image-guided neurosurgery to project images with regions of interest onto the patient's head, skull, or brain surface in real time. The aim of this study was to evaluate system accuracy and to perform the first intraoperative application. METHODS Images of segmented brain tumors in different localizations and sizes were created in 10 cases and were projected to a head phantom using a video projector. Registration was performed using 5 fiducial markers. After each registration, the distance of the 5 fiducial markers from the visualized tumor borders was measured on the virtual image and on the phantom. The difference was considered a projection error. Moreover, the image projection technique was intraoperatively applied in 5 patients and was compared with a standard navigation system. RESULTS Augmented reality visualization of the tumors succeeded in all cases. The mean time for registration was 3.8 minutes (range 2–7 minutes). The mean projection error was 0.8 ± 0.25 mm. There were no significant differences in accuracy according to the localization and size of the tumor. Clinical feasibility and reliability of the augmented reality system could be proved intraoperatively in 5 patients (projection error 1.2 ± 0.54 mm). CONCLUSIONS The augmented reality system is accurate and reliable for the intraoperative projection of images to the head, skull, and brain surface. The ergonomic advantage of this technique improves the planning of neurosurgical procedures and enables the surgeon to use direct visualization for image-guided neurosurgery.

Sign in / Sign up

Export Citation Format

Share Document