Cognition-based contrast adjustment using neural network based face recognition system

This paper introduces a contrast adjustment using neural-network based face recognition system. Parameter setting problem is generally solved by maximization or minimization of some objective evaluation functions such as correlation and statistical independence. However, the tuned filter output is not always adequate for face recognition system because filter and face recognition system are separately tuned using different criterions. It is also difficult to set an objective criterion for parameter setting because there are no correct solutions when we consider contrast adjustment problem. To handle such cases, we look to some subjective information such as face in the image, and directly employ facial recognition system as evaluation function for parameter setting. Experimental results show that cognition-based evaluation has a potential to adjust the image contrast.