Development of shape pattern recognition for FPGA-based object tracking system

This project presents Field Programmable Gate Array (FPGA) based solution for an object tracking system. The tracking design was applied in FPGA as rich information of FPGA offers one of the more powerful methods in object tracking. Object tracking system has a very useful application in many real world situations which require object detection. In general, a lot of different algorithms are used in various types of object tracking system such as the detection of handwriting, moving objects, barcodes etc. This project is focusing on shape detection to observe the real object image. This paper starts by describing the method of converting RGB image to binary signal using MATLAB. Then the binary signal is used as an input signal for the detection part. Detection algorithm used in the design is using a fixed method. The simulated waveform shows that the algorithm design can detect the shape correctly if and only if the input image is clear, located in the center of the image and does not flip or rotated. Verification done using FPGA board also proven that. But there are lot of improvement can be done to improve the design such as increasing the size of the binary signal for the input of the system.