Wenshuo Kang

Wenshuo Kang

Auto-generated profile, based on publicly available information

Wenshuo Kang

Shandong University of Science and Technology,Qingdao,China,266590

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2022 in Complexity
Verified
Journal Article
Open access
PDF file
DFDi Fan
Di Fan
WSWenxue Sun
Wenxue Sun
HZHuiyuan Zhao
Huiyuan Zhao
WKWenshuo Kang
Wenshuo Kang
CLChangzhi Lv
Changzhi Lv
In the Internet age, information security is threatened anytime and anywhere and the copyright protection of audio and video as well as the need for matching detection is increasingly strong. In view of this need, this paper proposes a zero‐watermarking algorithm for audio and video matching based on NSCT. The algorithm uses NSCT, DCT, SVD, and Schur decomposition to extract video features and audio features and generates zero‐watermark stream through synthesis, which is stored in a third‐party organization for detection and identification. The detection algorithm can obtain zero watermark from the audio and video to be tested and judge and locate tampering by comparing with the zero watermark of the third party. From the experimental results, this algorithm can not only detect whether the audio and video are mismatched due to tampering attacks but also locate the mismatched audio and video segments and protect the copyright.

26 May 2023
Verified
Conference Paper
Closed access
KWKaiWen Wu
KaiWen Wu
WKWenshuo Kang
Wenshuo Kang
Yiming LiYiming Li
Yiming Li
DFDi Fan
Di Fan
To solve the problem of whether workers wear helmets or not on the actual construction site, this paper proposes a helmet detection algorithm based on the integration of transformable attention and feature fusion. Since the hard hat is a small target, and the feature extraction is susceptible to background information interference and other factors, this paper proposes the TAM-YOLOv5 (Transformable Attention Mechanism-YOLOv5) detection model to detect whether workers wear the hard hat in the actual scene. Then, the detected head-and-shoulder Figure is binary classified in the Double branch-ResNet classification model. To determine whether the worker is wearing a helmet. The transformable attention mechanism is added to TAM-YOLOv5 based on the original YOLOv5 detection model, which will pay more attention to the feature information between channels and spatial positions. To solve the problem of inadequate feature extraction, feature fusion is added to the Double branch-ResNet model to enhance the expression ability of safety hat features. In addition, due to the diversity of safety helmets, this paper introduces mutual channel loss in terms of loss function to enhance the fine-grained classification between safety helmets. In this paper, the proposed algorithm is fully tested, and the results show that the algorithm has improved the recall and accuracy.