Optical Engineering · 2003 · 22 citations · 24 references
ArtsRadiologyBiomedical ImagingDiagnostic ImagingDiagnostic RadiologyImage AnalysisImage EnhancementImage ProcessingImage QualityImage Quality AssessmentMedical ImagingAdaptive Contrast EnhancementAdaptive Image ProcessingComputerized Image AnalysisContrast EnhancementContrast Enhancement AlgorithmsLocal Contrast EnhancementMammographic Image QualitySignal-to-noise Ratio
With a view of quantitatively evaluating image quality in mammographic systems from an automated computerized analysis of digitized mammographic phantom films, an image enhancement method for such images is presented. In this framework an adaptive neighborhood image processing technique with a contrast-enhancement function is used. The method consists of computing a local contrast around each pixel using a variable neighborhood whose size and shape depend on the statistical properties around the given pixel. The obtained image is then transformed into a new contrast image using several contrast-enhancement functions. Finally, an inverse contrast transform is applied to the previous image. To compare contrast-enhancement functions, images simulating objects similar to those observed in the phantom image, with various relative contrasts and signal-to-noise ratio (SNR) levels, are generated. Several parameters including the output-to-input SNR ratio and the mean squared error are used as comparison criteria. Results show that this process enhances features in the image with little enhancement of noise and that a trigonometric contrast-enhancement function performs best for studying phantom images.
24
Adaptive histogram equalization and its variations
Stephen M. Pizer, E. Philip Amburn, John D. Austin et al. · Computer Vision Graphics and Image Processing · 1987 · 3.9K citations
Radiography · 1997 · 438 citations
Natural Sciences, Physics, Radiology +12