Hybrid Featured in Face Recognition
Publish place: First National Conference on Advances in computer science and information retrieval approaches
Publish Year: 1392
نوع سند: مقاله کنفرانسی
زبان: English
View: 1,289
This Paper With 8 Page And PDF Format Ready To Download
- Certificate
- من نویسنده این مقاله هستم
این Paper در بخشهای موضوعی زیر دسته بندی شده است:
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
BPJ01_626
تاریخ نمایه سازی: 29 دی 1392
Abstract:
Face recognition is always looking for methods and systems which can perform face recognition same or better than human at acceptable precision and speed. Face recognition systems are mainly based on single biometric attributes. But through the time and the need to more precision and effectiveness and overcoming challenges and limitations of single biometric systems, composition in biometric systems found position in researches. These kinds of systems could be implemented in various levels such as composition in the level of sensor, attribute, guess and decision, that each could have different structure. One of the most important levels is composition in guess. Experiments show that the mentioned method is more effective in guess level. In present paper it is tried to show that with using the benefit of compound methods, face recognition as one of the most important biometrics would be more in attention. The proposal method is a compound method with parallel structure in guess level and has adaptive advantages of three extractor algorithm called PCA, LDA and Gabor operator for composition.To compound advantages a neural network of feed forward back propagation type is used. Experiments results on Feret, Ar and Yale datasets show that using suggested method could be helpful in decreasing extrinsic factors like brightness and changes causes from the position and the gesture on the different algorithms and improves the results. Also using different neural network, we perform detecting face status. Improved results on different algorithms in mentioned dataset are shown in table below.
Keywords:
Authors
Mohsen Zangian
Azad University Shahrood branch
Seyed Mohammad Reza Hashemi
Azad University, Qazvin branch eng
Vahid Rostami
Faculty Member of Azad University, Qazvin branch
مراجع و منابع این Paper:
لیست زیر مراجع و منابع استفاده شده در این Paper را نمایش می دهد. این مراجع به صورت کاملا ماشینی و بر اساس هوش مصنوعی استخراج شده اند و لذا ممکن است دارای اشکالاتی باشند که به مرور زمان دقت استخراج این محتوا افزایش می یابد. مراجعی که مقالات مربوط به آنها در سیویلیکا نمایه شده و پیدا شده اند، به خود Paper لینک شده اند :