The ANN based detector to remove random-valued impulse noise in images

• An artificial neural network (ANN) based method is proposed to detect random-valued impulse noise (RVIN) in images.
• The rank ordered absolute differences (ROAD) and the rank-ordered logarithmic difference (ROLD) values are used to train ANN.
• Six popular 512 × 512, 8-bit gray-level grayscale test images from the literature are chosen to test trained ANN.
• The performance of the proposed method is compared with ten different comparison filters.
• The proposed method provides significant improvement over comparison filters.

This paper presents an artificial neural network (ANN) based method to detect random-valued impulse noise (RVIN) in images. The proposed method employs the ANN to decide whether a pixel is corrupted or not with RVIN. The inputs of the ANN are the rank ordered absolute differences (ROAD) and the rank-ordered logarithmic difference (ROLD) values. After the detection process is completed, the corrupted pixels are restored by the edge-preserving regularization (EPR) method which allows edges and noise-free pixels to be preserved. The performance of the proposed method is evaluated on different test images and compared with ten different comparison filters from the literature. Simulation results indicate that the proposed method provides significant improvement over comparison filters especially for high noise densities.