This paper presents registration via embedded maps (REM), a deformable registration algorithm for images with varying topology. The algorithm represents 3-D images as 4-D manifolds in a Riemannian space (referred to as embedded maps). Registration is performed as a surface evolution matching one embedded map to another using a diffusion process. The approach differs from those existing in that it takes an a priori estimation of image regions where topological changes are present, for example lesions, and generates a dense vector field representing both the shape and intensity changes necessary to match the images. The algorithm outputs both a diffeomorphic deformation field and an intensity displacement which corrects the intensity difference caused by topological changes. Multiple sets of experiments are conducted on magnetic resonance imaging (MRI) with lesions from OASIS and ADNI datasets. These images are registered to either a brain template or images of healthy individuals. An exemplar case registering a template to an MRI with tumor is also given. The resulting deformation fields were compared with those obtained using diffeomorphic demons, where topological changes are not modeled. These sets of experiments demonstrate the efficacy of our proposed REM method for registration of brain MRI with severe topological differences.

Image-analysis methods play an important role in helping detect brain changes in and diagnosis of Alzheimer's Disease (AD). In this paper, we propose an automatic unsupervised classification approach to distinguish brain magnetic resonance (MR) images of AD patients from those of elderly normal controls. The symmetric log-domain diffeomorphic demons algorithm, with the properties of symmetry and invertibility, is used to compute the pair-wise registration, whose deformation field is then used to calculate the Riemannian distance between them. The spectral embedding algorithm is performed based on the Riemannian distance matrix to project images onto a low-dimensional space where each image is represented as a point and its neighboring points correspond to images of high anatomical similarity. Finally, the quick shift clustering method is employed in the embedded space to partition the dataset into subgroups. The experiments using the proposed method show very good performance for clustering images into AD and normal aging, using the Clinical Dementia Rating (CDR) scale as a comparison.

Towards the goal of achieving truly autonomous navigation for a surface vehicle in maritime environments, a critical task is to detect surrounding obstacles such as the shore, docks, and other boats. In this paper, we demonstrate a real-time vision-based mapping system which detects and localizes stationary obstacles using a single omnidirectional camera and navigational sensors (GPS and gyro). The main challenge of this work is to make mapping robust to a large number of outliers, which stem from waves and specular reflections on the surface of the water. To address this problem, a two-step robust outlier rejection method is proposed. Experimental results obtained in unstructured large-scale environments are presented and validated using topographic maps.

In this paper we present a vector-based 3D localization formula for a paracatadioptric omnistereo system. Based on vector representation, the performance of this stereo system is analyzed numerically, including the maximum detectable range and the uncertainty of 3D localization, with respect to the flexible stereo configuration of the system, positions of scene points, as well as errors in correspondence matching and errors in stereo configuration. The results of performance analysis are used to guide the trajectory of an autonomous surface vehicle (ASV), which is equipped with a paracatadioptric omnidirectional camera, in a map building application.

Shoreline detection plays an important role in vision based navigation for autonomous surface vehicles (ASVs). It is a challenging task because of the diversity in near-bank scenarios. In this paper, we present a two-stage algorithm to find the shoreline by employing multiple features. First, we classify images into two types: reflection-unidentifiable and reflection-identifiable. Based on this classification, images are further analyzed with suitable techniques respectively. In the reflection-unidentifiable case, the surface reflection is subtle and so the water region can be separated from land by an adaptive thresholding method. The points along the edge of the water region are then identified and the shoreline is estimated through a line-fitting technique. In the reflection-identifiable case, we aim to discriminate water regions from the land by means of a two-category region classifier. Images are oversegmented into small regions based on color homogeneity. Then the characteristic features of land-water scenes like symmetry and brightness are extracted and applied to classify a region into land or water categories. Experimental results show the efficacy of our approach and robustness in diverse situations

Multiscale analysis is often required in image processing applications because image features are optimally detected at different levels of resolution. With the advance of high-resolution 3D imaging, the extension of multiscale analysis to higher dimensions is necessary. This paper extends an existing 2D scale selection method, known as the minimum reliable scale, to 3D volumetric images. The method is applied to 3D boundary detection and is illustrated in examples from biomedical imaging. The experimental results show that the 3D scale selection improves the detection of edges over single scale operators using as few as three different scales.