Abstract
Background: De-noising is the main effect produced by image compression at low compression ratios, which alters only the noise parts of images. The difference between the original and manipulated images will be less than the statistical variation when pixel values are sampled from these two images. This study develops a method that determines the quality indication in images using compression.
Methods: We use the Kolmogorov-Smirnov (KS) two-sample test to determine if the pixel distribution in compressed images is statistically different from the original and then compare previous reports to find the conceived image compression level. Medical images are first compressed using JPEG2000 at various degrees. The KS test was then used to find whether the two datasets differ significantly. Four different window sizes with ten to one hundred thousand test positions are sampled independently from both the compressed and original images to determine their respective empirical distribution functions. Results: The rates for D over the critical value increased obviously with the increasing of image compression ratio. The rates of D values exceeding the critical values are independent of the numbers of test positions. The conceivable image compression ratio level in this work may be set at 10% below the rejection rate compared with previous reports. Conclusion: The results of this report prove that the KS test can be used to indicate the variation in image quality and results were proven equivalent to PSNR. The potential applications for this method include determining the optimal compression ratio for tele-radiology or image archiving.Keywords: Image compression, JPEG 2000, Kolmogorov-Smirnov test, PSNR, statistically lossless.
Graphical Abstract
Call for Papers in Thematic Issues
Intelligent Information Retrieval for Multispectral and Hyperspectral Imaging
Multispectral and hyperspectral imaging is an interdisciplinary research area, widely focuses on spectral, spatial, and temporal data. It offers a plethora of opportunities to efficiently analyze the vast area of the earth's surface. However, there are numerous factors such as dimensionality and size of the hyperspectral data, lesser training samples, ...read more
Machine and Deep Learning in Medical Image Diagnosis
Medical Image Processing regards a set of methodologies that have been developed over recent years with the purpose of improving medical image quality, improving medical data visualization, understanding, and assisting medical diagnosis, and so on. The application of machine and deep learning methods in sensing and imaging can potentially have ...read more
Quantum Machine Learning for Medical Data and Imaging
Quantum computing has promised a significant speedup in certain computationally intensive tasks that are intractable on classical computers. Researchers in quantum machine learning, such as those working in computer vision, image processing, biomedical analysis, and related topics, may play an important role in comprehending and working on complicated medical data, ...read more
13
1