Spot and Adjust Filter: a New Image Filter for Image Enhancement and Noise Reduction
Kai Yit Kok, Parvathy Rajendran, Aiffah Mohd Ali · International Journal on Engineering Applications (IREA) · 2020
A new image filter, Spot & Adjust filter, is proposed in this paper in order to remove noise from an image. It spots out the error pixels and recovers the intensity value using the absolute intensity difference and the average value of the local region. The performance of this filter is analyzed using seven types of image noise samples, which are Gaussian, Gamma, Rayleigh, Poisson, Salt and Pepper, Speckle and Uniform noise. Six types of image filters are analyzed in this work, including the Median, Minimum, Maximum, Arithmetic Mean, Geometric Mean, and Harmonic Mean filters. These image filters' performance is compared with the newly proposed filter in terms of Peak Signal to Noise Ratio (PSNR). The results have shown that the Spot & Adjust filter has excellent performance, especially in tackling Poisson noise and Salt & Pepper noise.