Local Radon Descriptors for Image Search
This work addresses content-based image retrieval, particularly for medical and general images, but is incremental as it builds on existing Radon transform methods by localizing the approach.
The paper tackled the problem of improving image retrieval by proposing Local Radon Descriptors (LRD) as a more discriminative alternative to global Radon descriptors, showing significant improvement in retrieval performance on datasets like IRMA and INRIA Holidays.
Radon transform and its inverse operation are important techniques in medical imaging tasks. Recently, there has been renewed interest in Radon transform for applications such as content-based medical image retrieval. However, all studies so far have used Radon transform as a global or quasi-global image descriptor by extracting projections of the whole image or large sub-images. This paper attempts to show that the dense sampling to generate the histogram of local Radon projections has a much higher discrimination capability than the global one. In this paper, we introduce Local Radon Descriptor (LRD) and apply it to the IRMA dataset, which contains 14,410 x-ray images as well as to the INRIA Holidays dataset with 1,990 images. Our results show significant improvement in retrieval performance by using LRD versus its global version. We also demonstrate that LRD can deliver results comparable to well-established descriptors like LBP and HOG.