EE 7730 Image Enhancement Bahadir K. Gunturk2 Image Enhancement The objective of image enhancement is to process an image so that the result is more suitable than the original image for a specific application. There are two main approaches: Image enhancement in spatial domain: Direct manipulation of pixels in an image Point processing: Change pixel intensities Spatial filtering Image enhancement in frequency domain: Modifying the Fourier transform of an image Bahadir K. Gunturk3 Image Enhancement by Point Processing Intensity Transformation Bahadir K. Gunturk4 Image Enhancement by Point Processing Contrast Stretching Bahadir K. Gunturk5 Image Enhancement by Point Processing Contrast Stretching Bahadir K. Gunturk6 Image Enhancement by Point Processing Intensity Transformation Bahadir K. Gunturk7 Image Enhancement by Point Processing Intensity Transformation Bahadir K. Gunturk8 Image Enhancement by Point Processing Intensity Transformation Bahadir K. Gunturk9 Image Enhancement by Point Processing Gray-Level Slicing Bahadir K. Gunturk10 Image Enhancement by Point Processing Bit Planes Bahadir K. Gunturk11 Image Enhancement by Point Processing Bit Planes Bahadir K. Gunturk12 Image Enhancement by Point Processing Histogram 0 255 Bahadir K. Gunturk13 Histogram Specification Intensity mapping Assume T(r) is single-valued and monotonically increasing. The original and transformed intensities can be characterized by their probability density functions (PDFs) Bahadir K. Gunturk14 Histogram Specification The relationship between the PDFs is Consider the mapping Cumulative distribution function of r Histogram equalization! Bahadir K. Gunturk15 Image Enhancement by Point Processing Histogram Equalization Bahadir K. Gunturk16 Image Enhancement by Point Processing Histogram Equalization Example Intensity Number of pixels Intensity Number of pixels Bahadir K. Gunturk17 Image Enhancement by Point Processing Histogram Equalization Bahadir K. Gunturk18 Histogram Specification Intensity mapping Assume T(r) is single-valued and monotonically increasing. The original and transformed intensities can be characterized by their probability density functions (PDFs) Bahadir K. Gunturk19 Histogram Specification The relationship between the PDFs is Consider the mapping Cumulative distribution function of r Histogram equalization! Bahadir K. Gunturk20 Histogram Specification Example Assume Equalization mapping is then found to be The inverse mapping is r must be between 0 and 1 Bahadir K. Gunturk21 Histogram Specification Example Check out the new PDF is Bahadir K. Gunturk22 Histogram Specification Assume we have a desired PDF Let the following be the equalization mappings Then, the desired mapping is Bahadir K. Gunturk23 Histogram Specification Bahadir K. Gunturk24 Histogram Specification Bahadir K. Gunturk25 Histogram Specification Bahadir K. Gunturk26 Histogram Specification Bahadir K. Gunturk27 Local Histogram Processing Histogram processing can be applied locally. Bahadir K. Gunturk28 Image Subtraction The background is subtracted out, the arteries appear bright. Bahadir K. Gunturk29 Image Averaging Original image Noise Corrupted image Assume n(x,y) a white noise with mean=0, and variance If we have a set of noisy images The noise variance in the average image is Bahadir K. Gunturk30 Image Averaging Bahadir K. Gunturk31 Spatial Filtering A low-pass filter A high-pass filter Bahadir K. Gunturk32 Spatial Filtering Median Filter Sort: ( ) Example Original signal: Noisy signal: Filter by [ 1 1 1]/3: Filter by 1x3 median filter: Bahadir K. Gunturk33 Spatial Filtering Median filters are nonlinear. Median filtering reduces noise without blurring edges and other sharp details. Median filtering is particularly effective when the noise pattern consists of strong, spikelike components. (Salt-and- pepper noise.) Bahadir K. Gunturk34 Spatial Filtering Original 3x3 averaging filter Salt&Pepper noise added 3x3 median filter Bahadir K. Gunturk35 Spatial Filtering Bahadir K. Gunturk36 Spatial Filtering Gradient Operators Averaging of pixels over a region tends to blur detail in an image. As averaging is analogous to integration, differentiation can be expected to have the opposite effect and thus sharpen an image. Gradient operators (first-order derivatives) are commonly used in image processing applications. Bahadir K. Gunturk37 Spatial Filtering Gradient Operators These are called the Sobel operators Bahadir K. Gunturk38 Spatial Filtering Laplacian Operators Laplacian operators are second-order derivatives. Bahadir K. Gunturk39 Spatial Filtering Bahadir K. Gunturk40 Spatial Filtering High-boost or high-frequency-emphasis filter Sharpens the image but does not remove the low-frequency components unlike high-pass filtering Bahadir K. Gunturk41 Spatial Filtering High-boost or high-frequency-emphasis filter High pass = Original Low pass High boost = (K)(Original) Low pass = (K-1)(Original) + Original Low pass = (K-1)(Original) + High pass When K=1, High boost = High pass When K>1, Part of the original is added back to the highpass result. Bahadir K. Gunturk42 Spatial Filtering A high-pass filter A high-boost filter Bahadir K. Gunturk43 Spatial Filtering High-boost or high-frequency-emphasis filter Bahadir K. Gunturk44 Spatial Filtering