Scale Space Smoothing, Image Feature Extraction and Bessel Filters
نویسندگان
چکیده
The Green function of Mumford-Shah functional in the absence of discontinuities is known to be a modified Bessel function of the second kind and zero degree. Such a Bessel function is regularized here and used as a filter for feature extraction. It is demonstrated in this paper that a Bessel filter does not follow the scale space smoothing property of bounded linear filters such as Gaussian filters. The features extracted by the Bessel filter are therefore scale invariant. Edges, blobs, and junctions are features considered here to show that the extracted features remain unchanged by varying the scale of a Bessel filter. The scale invariance property of Bessel filters for edges is analytically proved here. We conjecture that Bessel filters also enjoy this scale invariance property for other kinds of features. The experimental results presented here confirm our conjecture of the scale invariance property of the Bessel filters.
منابع مشابه
Why choosing advanced nonlinear scale space filtering for denoising and simplifying images?
Denoising, edge preserving smoothing and image simplification is of fundamental importance in a variety of image processing and computer vision applications during feature extraction and object detection procedures. The construction of an optimal pre-processing filtering tool for various corner/edge detection and segmentation tasks is still an open matter. Towards this end, here we argue that f...
متن کاملStochastic analysis of scale-space smoothing
In the high-level operations of computer vision it is taken for granted that image features have been reliably detected. This paper addresses the problem of feature extraction by scale-space methods. This paper is based on two key ideas: to investigate the stochastic properties of scale-space representations and to investigate the interplay between discrete and continuous images. These investig...
متن کاملProposals of Co-occurrence Frequency Image Based Filters
We have discussed that the co-occurrence frequency image (CFI) defined based on the co-occurrence frequency histogram of the gray value of an image has a potential to introduce a new scheme for image feature extraction. This paper proposes a couple of filters for image enhancements of such as sharpening and smoothing filters. These filters are very similar to but quite different from those whic...
متن کاملStochastic Analysis of Image Acquisition, Interpolation and Scale-space Smoothing
In the high-level operations of computer vision it is taken for granted that image features have been reliably detected. This paper addresses the problem of feature extraction by scale-space methods. There has been a strong development in scale-space theory and its applications to low-level vision in the last couple of years. Scale-space theory for continuous signals is on a firm theoretical ba...
متن کاملPose Invariant, Robust Feature Extraction From Range Data With a Modified Scale Space Approach
Feature-based Simultaneous Localization and Map building (SLAM) approaches require a robust method to extract position invariant landmarks from the surrounding environment. 2D laser range finders are currently one of the most common sensors used to obtain environmental information for mobile robot navigation due to their reliability, accuracy and low cost. However, the 2D laser scan data only g...
متن کامل