Loading...
Please wait, while we are loading the content...
Similar Documents
Recognizing altered face in and beyond visible spectrum
Content Provider | Indraprastha Institute of Information Technology, Delhi |
---|---|
Author | Nigam, Aastha |
Abstract | The performance of face recognition suffers due to many factors like pose, illumination, expres- sion, image quality, aging and disguise. This report addresses the problem of face recognition with altered/disguised faces. In case of disguise, important facial features are masked. Each light spectrum captures different characteristics of the face which can be used for recognition. Thermal imagery captures the heat patterns of the face. Hence, the effect of disguise is reduced in the face image. One of the main contributions of this report is the collection of disguised images of 75 participants in thermal, visible, and near-infrared spectrum. This report proposes an algorithm that encodes information from visible and thermal spectrum to recognize faces with disguise. The performance of this algorithm was evaluated on the IBVS Face Database and Notre Dame Face Database. This report also introduces the use of biometric patches in case of disguise. The performance of the proposed algorithm was compared with existing face recognition algorithms. The results show that biometric patches performed better than full faces. |
File Format | |
Language | English |
Access Restriction | Authorized |
Subject Keyword | Algorithms Image analysis Disguise Thermal Spectrum Visible Spectrum |
Content Type | Text |
Educational Degree | Bachelor of Technology (B.Tech.) |
Resource Type | Thesis |
Subject | Data processing & computer science |