A novel image denoising algorithm used for coal mine

作者：Lu Zhaolin^*; Hu Xiaojuan; Qian Jiansheng; Li Leida

来源：Journal of Information and Computational Science, 2012, 9(5): 1321-1327.

• Edge detectors
• Fidelity term
• Heat flow equations
• Hybrid model
• Image denoising algorithm
• Isotropic diffusion
• Scale parameter
• Second orders
• Smooth regions
• Staircase effect
• Total variation
• TV models

Because of the complicated circumstance, coal mine image contains a mass of noise. Based on the applications of Partial Differential Equations (PDE) to image processing, a novel method for coal mine image denoising is presented. The hybrid model combines Total Variation (TV) model and Heat Flow (HF) equation which are all second-order PDEs. The TV model preserves edges well, but easily generates staircase effect in smooth regions of the image. The heat flow equation is an isotropic diffusion function, which can avoid staircase effect, but cause edge blurring. A new edge detector is introduced to the presented model, which makes the model combine the two models and inherit their advantages. And the scale parameter in fidelity term of the model is improved to preserve the texture of the image. Compared experimental results distinctly demonstrate the superiority of the proposed model, in term of removing noise while preserving the edges and avoiding staircase effect.

• 出版日期2012
• 单位中国矿业大学（北京）