Residue feature analysis with empirical mode decomposition for mining spatial sequential patterns from serial remote sensing images

An extensive growth of serial remote sensing images paves the way for abundant data intended for the sequential spatial pattern determination in several fields like monitoring of agriculture, development of urban areas, and the vegetative area. However, conventional spatial sequential pattern mining is not applied efficiently or directly in the aspect of serial remote sensing images. Therefore, a residue feature analysis with empirical mode decomposition is proposed so as to enhance the spatial sequential pattern mining efficacy from the raster serial remote sensing images. At first, input images are being extracted by means of minima and maxima pattern by computing the mean of envelops and the intrinsic mode function components. If the intrinsic mode function condition is satisfied, then it is being subtracted from the original image; finally, the image is decomposed into many intrinsic mode functions and residue. The experimental outcomes attained indicate that the proposed strategy is proficient of mining spatial sequential pattern from the images of serial remote sensing. Though the support values of the patterns might not be attained accurately, the presented scheme guarantees that the whole patters are being extracted at lower consumption of time.