Logo Detection System for Live Videos
Image matching and recognition are the crux of OpenCV and they have a major part to play in everyday lives. In this paper, we present an armature for logo extraction in images and live videos. The wide range application of visual data from companies, institutions, individuals and social systems like “Flicker”, “YouTube” etc. is for wide distribution and sharing of images and live videos. Through this paper, the architecture of logo match and recognition which is important for brand advertisement and surveillance application is proposed. The system provides detail information regarding the location of logo in the live video and images. This process is carried out by using Speeded Up Robust Feature (SURF) algorithm. It uncovers either improper or unauthorized use of logos. Reference logos and test images are converted into binary form and their features are matched accordingly. The main aim of this project is to present an efficient and robust armature to discover as well as recognize logo images through the use of Computer Vision (OpenCV). The localization and recognition of logos from live video is a big challenge that has been undertaken in this research.