Textures within real images vary in brightness, contrast, scale and skew as imaging conditions change. To enable recognition of textures in real images, it is necessary to employ a similarity measure which is invariant to these properties. Furthermore, since textures often appear on undulating surfaces, such invariances must necessarily be local rather than global. Despite these requirements, it is only relatively recently that texture recognition algorithms with local scale and affine invariance properties have begun to be reported. Typically, they comprise detecting feature points followed by geometric normalization prior to description. We describe a method based on invariant combinations of linear filters. Unlike previous methods, we introduce a novel family of filters, which provide scale invariance, resulting in a texture description invariant to local changes in orientation, contrast and scale and robust to local skew. Significantly, the family of filters enable local scale invariants to be defined without using a scale selection principle or a large number of filters. A texture discrimination method based on the A2 similarity measure applied to histograms derived from our filter responses outperforms existing methods for retrieval and classification results for both the Brodatz textures and the UIUC database, which has been designed to require local invariance.
Abstract. Non-rigid registration of multimodal images is a challenging problem. One approach, maximization of mutual information, has been shown to be effective for registering certain image modalities and is currently considered the standard against which all other techniques are measured. In this paper, we propose an alternative representation of an image based on local phases rather than intensities; we then show how mutual information can be extended to this representation. Local phase acts as a description of local image structure, enabling mutual phase information to detect complex image relationships that are hard or impossible to detect using mutual intensity information. Typical results are presented, comparing the performance of phase and intensity mutual information methods on simulated MR and ultrasound images.
A method to determine the depth of buried localized radioactive contamination nonintrusively and nondestructively using principal component analysis is described. The γ-ray spectra from two radionuclides, cesium-137 and cobalt-60, have been analyzed to derive the two principal components that change most significantly as a result of varying the depth of the sources in a bespoke sand-filled phantom. The relationship between depth (d) and the angle (θ) between the first two principal component coefficients has been derived for both cases, viz. d(Φ) = x + y log(e) Φ where x and y are constants dependent on the shielding material and the γ-ray energy spectrum of the radioactivity in question, and φ is a function of θ. The technique enables the depth of a localized radioactive source to be determined nonintrusively in the range 5 to 50 mm with an accuracy of ±1 mm.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.