基于多核处理器的全方位图像展开优化

Optimization of Omnidirectional Image Unwrapping Based on Multi-core Processor

全方位图像展开算法运算量大,在当前的主流处理器中难以满足实时要求,对其仅做简单并行处理效果仍不理想。针对此问题,基于多核处理器改进双线性内插值法的全方位图像展开算法,对其实现并行处理。通过全方位图像展开实验,对展开时间、并行效率、渲染速度等进行分析和比较。实验结果表明,并行优化后,在E7200(双核2.53 GHz)下实时处理的分辨率能从640×480提高到1 024×768。

Omnidirectional image unwrapping algorithms are complex in calculation, so they are non-real time on present main processor. The direct parallelization effect is not satisfactory. To solve the problem, this paper improves omnidirectional image unwrapping algorithm based on the quadratic interpolation, realizes parallel processing on multiple-core processor, and analyzes the unwrapping time, parallel performance and rendering speed with the unwrapping experiment in different resolution. Experimental results indicate that it improves the real-time resolution 640x480 of the image unwrapping to 1 024×768 on Core 2 Duo E7200 2.53 GHz processor.