The D-MOT-10 FPGA IP core for motion detection in video stream images is a completed module intended to be used in vision systems for various applications (security systems, specialized systems, etc.) that are based on FPGA. The core represents a simple data exchange interface that ensures easy integration into various systems. The algorithms are perfectly suitable for motion detection of ground, surface and aerial objects of any type.

The core is connected to the interface controller of DDR memory via a standard interface (the controller is by Xilinx Company). The core reads current images of video sequence and records the results of work. The developer should ensure updating of image frames in the DDR memory according to multi-buffer principle (multiple buffering). Before operation is started, the core needs to be configured (configuration register values have to be written). The observed scene (in the images of the video sequence) should be fixed. Once initialized (switch-on), the core needs some time (2-3 minutes) to adapt to the scene. At this time, the core algorithms initialize filters and form a map of constantly moving objects (e.g. images of treetops swaying in the wind). Once all core filters adapted, all moving objects will be detected with a high degree of reliability.