Application Research of Night Intelligent Surveillance System in Residential Quarters

In recent years, Artificial Intelligence research has become a hot topic. However, Intelligent Surveillance based on vision is an important component part of Artificial Intelligence. The paper first extracts the parameters of exterior wall scene used Radon Transform and Polynomial Fitting; the Self-Adaptive Background Subtraction algorithm was developed to detect moving objects; the features of Orthonormal Gaussion-Hermite Moments were used for classification recognition of moving objects. Finally, an early warning system was established. The experiments show that the algorithm is effective.