Correlation-based multi-shape granulometry with application in porous silicon nanomaterial characterization

摘要: Image-based granulometry measures the size distribution of objects in an image granular material. Usually, algorithms based on mathematical morphology or edge detection are used for this task. We propose entirely new approach, using cross correlations with kernels different shapes and sizes. use pyramidal structure to accelerate multi-scale searching. The local maxima primary candidates centers objects. These candidate filtered criteria their intersection areas other Our technique spatially localizes each object its shape, rotation angle. This allows us measure many statistics (besides traditional distribution) e.g. shape spatial Experiments show that algorithm is greatly robust noise can detect even very faint noisy extract quantitative structural characteristics Scanning Electron Microscopy (SEM) images porous silicon layer. computes size, pores. relate these results fabrication process discuss rectangle formation mechanism. a reliable tool SEM processing.