Performance Analysis of Semantic Assisted Convolutional Neural Networks in Face Recognition
Journal Title: International Journal of Engineering and Science Invention - Year 2018, Vol 7, Issue 6
Abstract
In today's age of automation, face recognition is a vital component for authorization and security. It has received substantial attention from researchers in various fields of science such as biometrics and computer vision. Recently, deep convolutional neural networks have set a new trend in fields of face recognition by improving the state-of-the-art performance. By using deep neural networks, much more sophisticated and high level abstracted features can be learned automatically. In this paper, we propose a method for face recognition using semantic assisted convolutional neural network.(SCNN) . Our framework, referred to as SCNNs, incorporates explicit semantic information to automatically recover comprehensive face features. We demonstrate the performance of the SCNN model beginning from CNN basics. We then show results for face analysis using an external database with respect to existing systems basically NN and CNN. The results of the proposed approach were very encouraging and demonstrate superiority when compared with other techniques.
Authors and Affiliations
Kalyani D Sonawane, Dr. Sudhir D Sawarkar, Prof Archana Gulati
F4-TCNQ Concentration Dependent Capacitance-Voltage (C-V) and Conductivity-Voltage (G/w-V) Characteristics of the Au/P3HT:F4-TCNQ/N-Si (MPS) Schottky Barrier Diodes
In this work, the electrical properties of Au/P3HT:F4-TCNQ/n-Si metal-polymer-semiconductor (MPS) Schottky barrier diodes (SBDs) with the F4-TCNQ concentration values of 0.5% and 2%were investigated by capacitance-voltag...
Data Embedding an Image Using DCT Compression
This paper concerns with the hybrid technique that combines image compression and encryption to have an embedded image along with data that's secure and can be manipulated with a secure key. The compression of image can...
"Evaluation of the level of reading comprehension in university students applying the cloze test"
The purpose of this research is to evaluate the level of reading comprehension in university students of the School of Industrial Design of the Pontifical Catholic University of Ecuador-Ambato, using the Cloze test, cons...
Research of Physical Properties of Straw for Building Panels
The development of new thermal insulation materials needs to evaluate properties and structure of raw materials, technological factors that make influence on the thermal conductivity of material. One of the most promisin...
Next Generation App to Update Phone Number Automatically To All Contacts
With the rapidly advancing technology we are in the need of some unique applications that will help us to make our lives easier. The number of mobile Apps available has grown-up massively over the past few years. With th...