This letter proposes a new approach for feature extraction using steerable filters. This approach is based on the concept of orientation-energy histogram which yields the local direction of dominant orientation. The testing is carried out using a training set of 1000 and a set of 300 unknown 40 40 hand-written digits. As a result of the simulations, 92% correct recognition is provided.

In this letter, our proposed approach exploits the use of original and simplest Cellular Neural Network (CNN) for 2D Doubly Complementary (DC) Infinite Impulse Response (IIR) filter banks design. The properties of feedback and feedforward templates are studied for this purpose. Through some examples it is shown how generalizations of these templates can be used for DC IIR filter banks design. We modify Lagrangian function which is used for optimizing a low-pass filter design considering the constraint for stability of CNN. The brief conclusions with design examples that illustrate the proposed method and an image enhancement and restoration applications of designed filter banks are presented.