1. By Md.Naseem AshrafBE/5504/09IMAGE DEGRADATION&NOISEA Presentation for Digital Image ProcessingBirla Institute of Technology, Mesra, Extension Center - Patna
2. What is Image Degradation?Image degradation is said to occur when acertain image under goes loss of storedinformation either due to digitization orconversion (i.e algortithmic operations),decreasing visiual quality.
3. Image Degradation & RestorationModelThe initial image (source, f(x,y)) undergoes degradationdue to various operations, conversions and losses. Thisintroduces Noise. This Noisy Image is further restored viarestoration filters to make it visually acceptable for user.Degraded Image: = Degradation Function* Source + Noiseg(x,y) = h(x,y) * f(x,y) + n(x,y)
4. What is Noise?Image noise is random (not present in the object imaged)variation of brightness or color information in images, andis usually an aspect of electronic noise. It can be producedby the sensor and circuitry of a scanner or digital camera.Image noise can also originate in film grain and in theunavoidable shot noise of an ideal photon detector.Noisy Image Original Image
5. Some Important Noise ProbabilityDensity FunctionsGaussian NoiseSalt-and-Pepper (Impulse) NoisePoisson NoiseErlang (Gamma) NoiseExponential NoiseUniform Noise
6. Gaussian NoiseI = imread(eight.tif);J = imnoise(I,gaussian,0.02,0.1);figure, imshow(I)figure, imshow(J)Source Image Image with Gaussian NoiseMATLAB program for adding Gaussian Noise
7. Impulse (Salt and Pepper) Noise
8. MATLAB program for adding Impulse(Salt and Pepper) NoiseI = imread(eight.tif);J = imnoise(I,salt & pepper,0.02);figure, imshow(I)figure, imshow(J)Source Image Image with Salt & Pepper Noise
9. Poisson NoiseA random variable X that obeys a Poissondistribution takes on only nonnegative values;the probability that X = k is where λ is apositive parameter.
10. MATLAB program for adding PoissonNoiseI = imread(eight.tif);J = imnoise(I,poisson);figure, imshow(I)figure, imshow(J)Source Image Image with Salt & Pepper Noise
11. Erlang/Gamma NoiseOriginal Image Image with Gamma Noise
12. Exponential NoiseOriginal Image Image with Gamma Noise
13. Uniform NoiseOriginal Image Image with Gamma Noise