A Speckle Noise Removal Method
Autor: | Adnan Rashdi, Ayesha Saadia |
---|---|
Rok vydání: | 2017 |
Předmět: |
business.industry
Applied Mathematics 020208 electrical & electronic engineering ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Speckle noise Salt-and-pepper noise 02 engineering and technology Non-local means symbols.namesake Dark-frame subtraction Gaussian noise Computer Science::Computer Vision and Pattern Recognition Signal Processing 0202 electrical engineering electronic engineering information engineering Image noise Median filter symbols 020201 artificial intelligence & image processing Computer vision Artificial intelligence business Image restoration Mathematics |
Zdroj: | Circuits, Systems, and Signal Processing. 37:2639-2650 |
ISSN: | 1531-5878 0278-081X |
Popis: | During formation of echocardiographic images, speckle noise is introduced, which diminishes important information present in an image and effects physician's capability to interpret image correctly. In the literature, many techniques have been proposed to remove unwanted noise from the image. In this paper, an intelligent denoising algorithm for echocardiographic images has been proposed, which first divides input image into different regions, namely smooth, texture and edge, using coefficient of variation. Fuzzy logic is used to draw boundaries between these image regions. Average filter and fractional integral filters are deployed to denoise pixels of various regions. Selection of filter depends on the characteristics of a region. The proposed technique improves quality of denoised image by suppressing maximum noise and producing no artifacts. Simulation results show superiority of proposed methodology over state-of-the-art existing methodologies, visually and using quantitative measures i.e. mean square error, peak signal to noise, edge preservation index, correlation coefficient and structure similarity. |
Databáze: | OpenAIRE |
Externí odkaz: |