Abstract
In the process of traditional urban enforcement supervision, the illegal building detection are mainly based on the inspection of the naked eye. Due to factors such as fatigue and the environment, the detection efficiency is low and it is error-prone. This article focuses on the automatic detection algorithm of fixed-point monitoring video images as samples, and performs background separation, match, repair and other pre-processing on the sampled video images of different phases. For the weather blocking the surrounding environment and many other interference factors, the morphological operators are used to filter and extract clear areas of change, then the HU moments and template matching evaluation function are combined to clean the changed sub-regions, and finally the spatial characteristics are studied to detect the building. The experimental results show that this scheme not only improves the computational efficiency but also ensures the recognition rate.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have