Abstract:The template matching is a basic operation that is commonly met in image processing and computer vision. It has been used as an basic and efficient method for filtering, edge detection, object recognition and image matching. Template matching can be divided into non-cyclical template matching and cyclical template matching. Suppose the image and template sizes are N×N and M×M, respectively. It is clear that cyclical template matching can be done in O(N 2M 2) time if we use traditional algorithms in a uniprocessor. Thus it can be seen that template matching, just as most image processing and computer vision, is intensive problems in computation and should exert parallel processing on the elements of the image. In this paper, template matching algorithms used on LS MPP parallel array computer that we have studied and are discussed in detail. Firstly, the parallel computation principles of template matching algorithms are discussed. Secondly, based on the application of LS MPP parallel array computer, we design data parallel moving algorithms that is needed in the course of parallel operation and an image template matching parallel algorithm that is suited to K-ary 2-cube network SIMD computer with non-restricted local register. Finally, through analyzing the application limitation of the image template matching parallel algorithm of non-restricted local register, this paper presents an image template matching parallel algorithm for K-ary 2-cube network SIMD computer with restricted local registers, and the algorithm is implemented on LS MPP. For an N×N image and M×M template (K=N, MN), both the communication complexity and the computation complexity of the algorithm of restricted local register are all O(M 2). Obviously,the parallel algorithm is an optimal speed-up algorithm.

An Efficient Algorithm for Complete Euclidean Distance Transform on Mesh-Connected Simd

A Constant-Time Algorithm for Computing the Euclidean Distance Transform on Reconfigurable Meshes

A Fast Algorithm for Complete Euclidean Distance Transformation Based on Contour Scanning

Efficient Euclidean Distance Transform Algorithm of Binary Images in Arbitrary Dimensions.

A Multi-Threaded Parallel Algorithm for Quality Improvement of Tetrahedral Meshes

Efficient Euclidean distance transform using perpendicular bisector segmentation

Algorithm of complete Euclidean distance transformation based on contour hierarchical scanning

Parallel Collision Detection Between Deformable Objects Using SIMD Instructions

A Fast Algorithm for Image Euclidean Distance

Efficient parallel algorithms for distance maps of 2D binary images using an optical bus

A novel high performance discrete cosine transform algorithm based on the stochastic computation

Mccd Multi-Core Collision Detection Between Deformable Models Using Front-Based Decomposition

Faster Line Detection Algorithms on Enhanced Mesh Connected Arrays

Optimized Space Decomposition Algorithm for Parallel Continuous Collision Detection

Parallel Computing of A Digital Hologram and Particle Searching for Microdigital-Holographic Particle-Tracking Velocimetry

Simd Massively Parallel Processing System For Real-Time Image Processing

A parallel algorithm for the computation of two-dimensional image moments on SIMD-MCC

Parallel Compact Differential Evolution for Optimization Applied to Image Segmentation

A 250mhz Optimized Distributed Architecture Of 2d 8x8 Dct

Parallel Algorithm for Image Template Matching on A K-ary 2-Cube Network SIMD Computer

A Progressive Approach for Computing the Earth Mover's Distance