Gestalt principle based multipart object and object group detection on FPGA

Finding useful information in real world scenes is very important for many scene understanding tasks. Signs, scripts, information panels, and logos typically stand out from their environment for a human observer, but locating them seems to be inherently dependent on the context. We present a connected component labeling and a grouping algorithm fitted for FPGAs to locate potentially interesting areas and examine whether their content is worth further analysis. Experimental results are shown for real word images.