Medical Image Comparison Using Image Features  
Author Anik Roy


Co-Author(s) Partha Sarathi Mukherjee


Abstract Image monitoring is an important research problem in the areas of medical science, satellite imaging, security analysis, and so forth. In the literature, many of the image monitoring techniques are intensity-based, and they incorrectly signal a change when there is an alteration in the contrast of the background and foreground whereas the boundaries of the image objects remain unchanged. This is because the techniques are based solely on the intensities which do not take into account other important image features, such as edges. To accomplish this, we need to compare two images, and specifically, the edge location curves (JLC) of two images. This paper proposes a testing procedure to compare two grayscale images based on jump location curves (JLC). A number of numerical studies show that such an approach is useful in many applications.


Keywords Edge-based image comparison, Hausdorff distance, Hypothesis testing
    Article #:  DSBFI23-144
Proceedings of 2nd ISSAT International Conference on Data Science in Business, Finance and Industry
January 8-10, 2023 - Da Nang, Vietnam