Navigating Weighted Regions with Scattered Skinny Tetrahedra

We propose an algorithm for finding a [Formula: see text]-approximate shortest path through a weighted 3D simplicial complex [Formula: see text]. The weights are integers from the range [Formula: see text] and the vertices have integral coordinates. Let [Formula: see text] be the largest vertex coordinate magnitude, and let [Formula: see text] be the number of tetrahedra in [Formula: see text]. Let [Formula: see text] be some arbitrary constant. Let [Formula: see text] be the size of the largest connected component of tetrahedra whose aspect ratios exceed [Formula: see text]. There exists a constant [Formula: see text] dependent on [Formula: see text] but independent of [Formula: see text] such that if [Formula: see text], the running time of our algorithm is polynomial in [Formula: see text], [Formula: see text] and [Formula: see text]. If [Formula: see text], the running time reduces to [Formula: see text].

Research of Infrared Thermal Imagery Segmentation Technology Based on Visible Light Image

To solve the problem of infrared target recognition, byusing the complementarity of visible-light image and infrared-thermal imagery, thisarticle presents a kind of infrared thermal imagery segmentation technology. Segmentingthe target edges of visible-light images, and superimposing the edge on thecorresponding infrared thermal imagery, then segmenting the infrared thermalimagery by the improved weighted regions growing algorithm. After the testabout relevant parameters of the infraredthermal imagery, found that contrast enhancement and entropy increase, witchmaking it easy to split and recognize, and human eye subjective judgment isalso much easier. It put forward a new research method about infrared targetrecognition

APPROXIMATE SHORTEST HOMOTOPIC PATHS IN WEIGHTED REGIONS

A path P between two points s and t in a polygonal subdivision [Formula: see text] with obstacles and weighted regions defines a class of paths that can be deformed to P without passing over any obstacle. We present the first algorithm that, given P and a relative error tolerance ε ϵ (0, 1), computes a path from this class with cost at most 1 + ε times the optimum. The running time is [Formula: see text], where k is the number of segments in P and h and n are the numbers of obstacles and vertices in [Formula: see text], respectively. The constant in the running time of our algorithm depends on some geometric parameters and the ratio of the maximum region weight to the minimum region weight.