Image Processing Hardware Algorithm for Monitoring Environment

Abstract

Recently, object recognition applications using cameras in various devices have been developed and it is expected to be used in many fields in the future. In this paper, we designed SURF based hardware to detect forest fire. We designed the hardware and used a parallel processing structure to enable real-time processing. We also proposed a method to reduce memory usage to reduce the area of hardware. We have designed the hardware in HDL and verified its operation on Modelsim simulation comparing to the result of Matlab software. This design enables real-time monitoring of environments such as fire detection and air pollution.