A linked list run-length-based single-pass connected component analysis for real-time embedded hardware

摘要: Conventional connected component analysis (CCA) algorithms render a slow performance in real-time embedded applications due to multiple passes resolve label equivalences. As this fundamental task becomes crucial for stream processing, single-pass were introduced enable stream-oriented hardware design. However, most CCA the literature inhibit maximum streaming throughput as additional time such horizontal blanking period is required equivalence. This paper proposes novel algorithm, using combination of linked list and run-length-based techniques equivalences well extracting object features single raster scan. The proposed algorithm involves recycling scheme which attains low memory requirement Experimental results show implementation achieves one cycle per pixel surpasses memory-efficient state-of-the-art work up 25 % reduction usage $$7680\times 4320$$  pixels image.