Patch Neighborhood Research Articles

LESC: Superpixel cut‐based local expansion for accurate stereo matching

The rapid estimation of the accurate disparity between pixels is the goal of stereo matching. However, it is very difficult for the 3D labels-based methods due to huge search space of 3D labels, especially for high-resolution images. In this paper, a novel superpixel cut-based method is proposed, in an attempt to get the accurate disparity map efficiently, including the multi-layer superpixel optimization and iteractive local α-expansion in parallel. As for the multi-layer superpixel optimization, feature point optimization is designed to get accurate candidate labels that are set for most pixels using non-local cost aggregation strategy and update per-pixel labels of the corresponding superpixels from the candidate label sets on the small-size superpixel layer, and then update the middle to large-size superpixel layers progressively using non-local cost aggregation strategy. In order to provide more prior information to identify weak texture and textureless regions in non-local cost aggregation, the weight combination of “intensity + gradient + binary image” is proposed for constructing an optimal minimum spanning tree (MST) to calculate the aggregated matching cost and obtain the labels of minimum aggregated matching cost. Moreover, the local patch surrounding the corresponding superpixels is designed to accelerate superpixel optimization in parallel, and a neighborhood structure is presented to optimize the algorithm in this study, including superpixel neighborhood and patch neighborhood. As for the iteractive local α-expansion, three layers of patch structure corresponding to the superpixel neighborhood structure is proposed for optimizing the algorithm in this study. The experimental results show that higher accuracy could be achieved via the method in this study compared with some known state-of-the-art stereo methods on KITTI 2015 and Middlebury benchmark V3, which are the standard benchmarks for testing the stereo matching methods.

Color and Laser Data as a Complementary Approach for Heritage Documentation

Heritage recording has received much attention and benefits from recent developments in the field of range and imaging sensors. While these methods have often been viewed as two different methodologies, data integration can achieve different products, which are not always found in a single technique. Data integration in this paper can be divided into two levels: laser scanner data aided by photogrammetry and photogrammetry aided by scanner data. At the first level, superior radiometric information, mobility and accessibility of imagery can be actively used to add texture information and allow for new possibilities in terms of data interpretation and completeness of complex site documentation. In the second level, true orthophoto is generated based on laser data, the results are rectified images with a uniform scale representing all objects at their planimetric position. The proposed approaches enable flexible data fusion and allow images to be taken at an optimum time and position for radiometric information. Data fusion usually involves serious distortions in the form of a double mapping of occluded objects that affect the product quality. In order to enhance the efficiency of visibility analysis in complex structures, a proposed visibility algorithm is implemented into the developed methods of texture mapping and true orthophoto generation. The algorithm filters occluded areas based on a patch processing using a grid square unit set around the projected vertices. The depth of the mapped triangular vertices within the patch neighborhood is calculated to assign the visible one. In this contribution, experimental results from different historical sites in Jordan are presented as a validation of the proposed algorithms. Algorithms show satisfactory performance in terms of completeness and correctness of occlusion detection and spectral information mapping. The results indicate that hybrid methods could be used efficiently in the representation of heritage structures.

Testing for the relative contribution of patch neighborhood, patch internal structure, and presence of predators and competitor species in determining distribution patterns of rodents in a fragmented landscape

Factors affecting the distribution of species in fragmented landscapes, and their relative importance, are often unclear. Few studies have deliberately compared various hypotheses in relatively controlled conditions. Moreover, most studies to date have not incorporated false absences in their modeling. Following a multiple hypotheses testing framework, we tested the relative role of landscape structure, patch and neighborhood configuration, patch internal structure, presence or absence of predators, and presence or absence and abundance of potential competitor species on the occupancy patterns of three rodent species ( Myodes glareolus (Schreber, 1780), Apodemus flavicollis (Melchior, 1834), Apodemus sylvaticus (L., 1758)) in a highly fragmented landscape in Tuscany, central Italy. Distribution patterns and occupancy dynamics of the three rodent species were affected by patch and patch neighborhood characteristics, as well as the vicinity of other patches occupied by the species. However, we found no relevant effect of the surrounding landscape structure, patch internal structure, predator patch use, and abundance or distribution of potential competitor species. The observed spatial correlation between populations and the evidence of colonization or extinction events suggest that a metapopulation approach could provide a good framework for understanding the long-term dynamics of these populations.

Spatial adjacencies and interactions: Neighborhood mosaics for landscape ecological planning

Ecological planning routinely falls short in considering the spatial configuration of land-use types and intensities even though it is well known that unfavorable configurations can lead to undesirable effects on humans and nature. The neighborhood mosaic is proposed as a promising unit for studying, planning, and managing spatial configurations. The neighborhood mosaic is a local assemblage of landscape elements linked together by strong interaction. Three aspects of the neighborhood mosaic concept are crucial: patch adjacency, patch-and-matrix pattern, and patch neighborhood. Examples from landscape ecology and planning are used to describe each aspect. A list of questions is provided to help planners to determine the appropriate facet for a given planning situation and to define the neighborhood mosaic. Based on that, a list of strategies is provided to improve the planner's ability to create spatial configurations that take into account the effects land uses exert onto each other. The spatial concept of the neighborhood mosaic is a tool for analysis as well as a design concept with the potential to foster the integration of landscape ecological knowledge into planning. There is much potential for developing the presented qualitative approach into a quantitative one.

Surface parametrization and curvature measurement of arbitrary 3-D objects: five practical methods

Curvature sampling of arbitrary, fully described 3-D objects (e.g. tomographic medical images) is difficult because of surface patch parameterization problems. Five practical solutions are presented and characterized-the Sander-Zucker approach, two novel methods based on direct surface mapping, a piecewise linear manifold technique, and a turtle geometry method. One of the new methods, called the cross patch (CP) method, is shown to be very fast, robust in the presence of noise, and is always based on a proper surface parameterization, provided the perturbations of the surface over the patch neighborhood are isotropically distributed. >