This preview shows pages 1–6. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Restoration in the Presence of Noise Only – Spatial Filtering COMP344 COMP344 Restoration in the Presence of Noise Only – Spatial Filtering Restoration in the Presence of Noise Only g ( x , y ) = h ( x , y ) * f ( x , y ) + η ( x , y ) When noise is the only image degradation present in an image g ( x , y ) = f ( x , y ) + η ( x , y ) G ( u , v ) = F ( u , v ) + N ( u , v ) In case of periodic noise , it is usually possible to estimate N ( u , v ) from the spectrum of G ( u , v ) and subtract it to obtain an estimate of the original image f ( x , y ) In general, estimating the noise terms η ( x , y ) is unreasonable, so subtracting them from g ( x , y ) is impossible Spatial filtering is the method of choice in situations when only additive noise is present COMP344 Restoration in the Presence of Noise Only – Spatial Filtering Spatial Filter: Mean Filter Mean filters Arithmetic mean filter Geometric mean filter Harmonic mean filter Contraharmonic mean filter Common to all mean filters is that they operate on a rectangular subimage of size m × n centred at point ( x , y ) COMP344 Restoration in the Presence of Noise Only – Spatial Filtering Arithmetic Mean Filter Arithmetic mean given { a 1 , a 2 , . . . , a n } 1 n ∑ n i =1 a i Arithmetic mean filter computes the average value of the corrupted image g ( x , y ) in the area defined by S xy ˆ f ( x , y ) = 1 mn ∑ ( s , t ) ∈ Sxy g ( s , t ) smooths local variance of the image can also be implemented by using a convolution mask with all the coefficients equal to 1 / ( mn ) COMP344 Restoration in the Presence of Noise Only – Spatial Filtering Geometric Mean Filter Geometric mean Q n i =1 a i 1 / n = n √ a 1 · a 2 . . . a n Example Geometric mean of 2 and 8: 4 Geometric mean filter computes the product of the pixels of the corrupted image g ( x , y ) in the area defined by S xy . Then, it raises the product to the power 1 / ( mn ) ˆ f ( x , y ) = h Q ( s , t ) ∈ Sxy g ( s , t ) i 1 mn achieves smoothing comparable to the arithmetic mean filter, but it tends to lose less image detail in the process...
View
Full
Document
 Fall '09
 AlbertK.Wong

Click to edit the document details