Abstract
Background: Edge preserving filters are widely preferred for medical image denoising as they do not degrade the morphological edges during smoothing. Even though, the operational parameters of these filters have crucial influence on their performance, the parameters are selected subjectively. The optimum values of the operational parameters can be selected objectively with the help of edge quality indices. The available edge quality indices either do not comply with the subjective quality ratings or they are prone to noise level.
Objectives: (a) To formulate an edge preservation metric which has good correlation with subjective fidelity ratings and is robust to noise (b) To demonstrate an objective method for the selection of optimum values of the operational parameters of the non-linear spatial filters using the newly formulated edge preservation metric. Methods: Pratt’s Figure of Merit (PFOM) between the binary edge maps of the original and restored images, extracted via gradient based threshold, is used as the measure of extent to which edges are preserved during restoration. Magnetic Resonance (MR) images filtered by anisotropic diffusion for different values of number of iterations are used as ground truth images. The PFOM is compared with existing edge quality indices in terms of robustness to noise and correlation with subjective fidelity ratings. Results: PFOM exhibits a correlation of 0.9998 with the subjective edge quality rating which is only 0.9802 for Edge Preservation Index (EPI). Conclusion: The proposed index is robust to noise level and useful for optimizing the performance of non-linear spatial filters.Keywords: Edge preservation index, Edge quality index, Image quality analysis, Pratt’s figure of merit, non-linear restoration.
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
15
3