Search

Video is a rich media widely used in many of our daily life applications like education, entertainment, surveillance, etc. In order to retrieve rapidly, it is necessary to establish digital archive for storing these videos. However, it is not realistic to store vast amounts of video data into digital archive artificially. This paper proposes a new method for the task of video digital archive management by employing scene recognition technology based on extreme learning machine (ELM). This paper only focuses on scene recognition technology which is the key step of digital video archive management. Dense scale invariant feature transform (dense SIFT) features are used as features in this proposed method. The 15-Scenes dataset with more than 4000 images is used. Experimental results have shown that this proposed method achieves not only high recognition accuracy but also extremely low computational cost.

Detecting vehicles in video is a challenging problem owing to the motion of vehicles, the camera and the background and to variations of speed. This paper proposes a classifier based supervised method to detect moving vehicles from a moving camera. Dense scale invariant feature transform (dense SIFT) descriptors are used as features to describe the pattern of the object. And Extreme Learning Machine provides excellent generalization performance at fast speed. Our sample images taken by a camera in helicopter include 2000 images. Experiment results shown that this proposed method has not only good overall performance but also low computational cost.