Real-time hardware–software embedded vision system for ITS smart camera implemented in Zynq SoC

摘要: The article demonstrates the usefulness of heterogeneous System on Chip (SoC) devices in smart cameras used intelligent transportation systems (ITS). In a compact, energy efficient system following exemplary algorithms were implemented: vehicle queue length estimation, detection, counting and speed estimation (using multiple virtual detection lines), as well type (local binary features SVM classifier) colour (k-means classifier YCbCr colourspace analysis) recognition. solution exploits hardware–software architecture, i.e. combination reconfigurable resources ARM processor. Most modules implemented hardware, using Verilog HDL, taking full advantage possible parallelization pipeline, which allowed to obtain real-time image processing. processor is responsible for executing some parts algorithm, high-level processing analysis, communication with external (e.g. traffic lights controllers). demonstrated results indicate that modern SoC are very interesting platform advanced ITS other embedded processing, analysis recognition applications.