ECOC-Based Training of Neural Networks for Face Recognition
This is an incremental improvement for face recognition systems, enhancing classification reliability in a specific domain.
The paper tackled improving face recognition accuracy by applying Error Correcting Output Codes (ECOC) to train feed-forward neural networks, achieving high reliability with a defined rejection scheme on the Yale database.
Error Correcting Output Codes, ECOC, is an output representation method capable of discovering some of the errors produced in classification tasks. This paper describes the application of ECOC to the training of feed forward neural networks, FFNN, for improving the overall accuracy of classification systems. Indeed, to improve the generalization of FFNN classifiers, this paper proposes an ECOC-Based training method for Neural Networks that use ECOC as the output representation, and adopts the traditional Back-Propagation algorithm, BP, to adjust weights of the network. Experimental results for face recognition problem on Yale database demonstrate the effectiveness of our method. With a rejection scheme defined by a simple robustness rate, high reliability is achieved in this application.