With the spatial, spectral and temporal resolutions of remote sensing data increasing, the computing efficiency becomes one of bottlenecks for remote sensing image data processing, especially for that with time response requirements. In this paper, towards the aerosol optical depth retrieval application from moderate resolution imaging spectroradiometer data, taking the time-consuming interpolation pre-processing as the study case which includes the inverse distance weighted and the bilinear interpolation methods, parallel computing methods were designed and implemented based on OpenMP programming model. The parallel runtime were measured and analyzed from the aspects of different image data size and threads. Experimental results show that the multi-core parallelization based on OpenMP programming model can efficiently reduce runtime, and offer suggestions for efficient desktop solutions with the evolving multicore architectures.