A Large-Batch Orthorectification Generation Method Based on Adaptive GPU Thread Parameters and Parallel Calculation

Orthorectification reflects a large amount of real and objective information, such as the characteristics images geometric accuracy maps. Conducting batch orthorectification is process with high time cost owing to pixelwise correction each image. A common approach use graphics processing unit (GPU) parallel computing accelerate processing. However, most existing GPU acceleration studies have adopted experimental testing methods determine thread parameters, which are inapplicable different GPUs affect performance. We put forward an adaptive calculation method for parameters based on performance by simultaneously blocking image automatically according memory space. used 112 ZY-3 test compare it general GPU. The results show following: first, single image, presented in this article 43.22% higher than that GPU, 34.41 times faster central unit. result automatic was same artificial blocking. Second, four indicated all exhibited significant speed boost. Third, large-batch images, 32.6% provides optimization strategy orthorectification.