Search

In some complex scenes of tracking, the traditional color feature based target tracking algorithm may have the unstable tracking. To solve this problem, a color and texture feature based target tracking algorithm is proposed. The texture feature is added in the traditional color Mean-shift based target tracking algorithm, and both the texture and the color features are extracted together. According to the sequence principle, the traditional color based mean drift tracking algorithm is still used to search the new position of the target, but, in each iteration process, the texture feature is used to search the optimal target position which has the largest similar feature. And the eight neighborhood search method(8areas with the same size are expanded around the candidate area)is applied to solve the problem of partial occlusion. The comparative experiments show that the algorithm has better robustness and realtime performance in complex scenes. [ABSTRACT FROM AUTHOR]

Integrity protection is an important content of computer security. Most of security OSes supported integrity protection mechanism, but integrity also may be destructed, and the protection mechanism is not flexible enough. Based on the principle of integrity protection, FIC (Flexible Integrity Control) policy is proposed and implemented under LSM (Linux Security Module). Integrity level and integrity auxiliary level is defined. FIC defines many rules including access control rules, process re-label rules and new objects labeled rules, FIC policy can protect system's integrity and process executions flexible integrity protection. Implementation effect is analyzed. Scalable requirement is pointed out. [ABSTRACT FROM AUTHOR]

Social media Key Opinion Leader (KOL) is very popular with advertisers because of their excellent business value. However, with the low entry threshold of the KOL industry and data fraud, advertisers are unable to find a KOL that matches their own brand quickly. Based on the above background, this paper studies the video released by KOL on social platforms, analyzes the dynamic theme of the barrage text in the video, and describes the change of the theme of the barrage over time. At the same time, a convolutional neural network model is used to perform sentiment analysis on the barrage text of the video containing the advertisement, and further analyze the audience's emotional polarity to the situation that the video released by KOL containing the advertisement. The experimental results show that the proposed KOL analysis method evaluates the commercial value of KOL more comprehensively and specifically, helping advertisers find a suitable KOL efficiently. [ABSTRACT FROM AUTHOR]

A pedestrian detection method is proposed based on the fusion of HSV color space feature and LBP feature. HSV feature is global which can simply describe the global distribution of colors in an image while LBP feature is a good description of the spatial structure of a local image, therefore the proposed method takes both the global features and the local features into account, which has the advantages of low dimension and quick computing speed. In Matlab, Adaboost classifier is used to carry out the experiments. Results show that, compared with the classic Gradient Direction Histogram (HOG) feature, LBP feature, Pyramid Histogram of Oriented Gradients (PHOG) feature and HOG-LBP feature, the proposed method is of high recognition rate. [ABSTRACT FROM AUTHOR]