6533b82dfe1ef96bd1291ca7

RESEARCH PRODUCT

Patch-Based Image Denoising Model for Mixed Gaussian Impulse Noise Using L1 Norm

Munesh Chandra TrivediPuneet Kumar GoyalVikash Kumar SinghMohan Lal KolhebManish Shrimali

subject

Mathematical optimizationbusiness.industryComputer scienceGaussianComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONTransmission mediumImpulse (physics)Non-local meansImpulse noiseComputer graphicssymbols.namesakeGaussian noiseComputer Science::Computer Vision and Pattern RecognitionsymbolsComputer visionArtificial intelligencebusinessImage restoration

description

Image denoising is the classes of technique used to free the image form the noise. The noise in the image may be added during the observation process due to the improper setting of the camera lance, low-resolution camera, cheap, and low-quality sensors, etc. Noise in the image may also be added during the image restoration, image transmission through the transmission media. To obtain required information from image, image must be noise free, i.e., high-frequency details must be present in the image. There are number of applications where image denoising is needed such as remote location detection, computer vision, computer graphics, video surveillance, etc. In last two decades, numbers of models have been published for image denoising. In this research work, we proposed patch-based image denoising model for mixed impulse, Gaussian noise using L1 norm. Mat lab 2014a on the Intel i5 with 4 GB RAM platform is used to simulate the proposed model. Simulation results show the effectiveness of our proposed model for image denoising as compared to state-of-the-art methods.

yearjournalcountryeditionlanguage
2017-10-24
https://doi.org/10.1007/978-981-10-5523-2_8