Real time realization of basic image processing algorithms on FPGA

Image processing, which has been one of the areas of intensive work in recent years, has been successfully used in many engineering fields. Image processing includes processes whose inputs and outputs are images, as well as processes that extract attributes from these images. Examples of these processes are segmentation, morphological operations, edge finding, corner finding and filtering. The pictures or images used for these operations consist of pixels. 307,200 pieces of data need to be calculated in order to be able to process an image of 640x480 pixels, which is transferred via a VGA (Video Graphics Array). As image quality or size increases, the transaction volume also increases. In order to carry out these transactions in real time, digital platforms that can perform high speed and parallel transactions are required. In this study presented, FAST (Features from Accelerated Segment Test) and Harris corner detection algorithms, Sobel algorithm from edge detection algorithms, spreading and erosion methods from morphological processing algorithms, and color changing algorithm FPGA (Field Programmable It is designed to run on Gate Array (Field Programmable Gate Arrays) chips in real time. Xilinx Vivado Design Suite HLx was used in the designed real time image processing algorithms. Real-time images were taken from the camera via HDMI (High Definition Multimedia Interface). VHDL (Very High Speed ​​Integrated Circuit Hardware Description Language) was used to process the real-time image data received on FPGA. All the designs were implemented on the Xilinx Zybo Z7-20 card. Image data obtained from the results of FPGA-based designs were transferred to the monitor via HDMI. The results obtained from real-time image processing algorithms designed to work on Xilinx Zybo Z7-20 FPGA board are presented.