D. Chen, X. Zhang, and K.P. Chan (PRC)
off-line signature verification, biometric, adaptive, weighting
In this paper we propose an adaptive weighting algo rithm to improve the recognition accuracy in off-line sig nature verification. The algorithm uses expert examiners' judgment for confused edge samples. Meanwhile, the judg ment is fed back to the verification system to boost the performance. We give different weights to different sam ples using in k-NN method so that higher accuracy can be achieved. This method is not only useful in area of signa ture verification but also a general and promising approach in all kinds of verification and recognition problems.
Important Links:
Go Back
