http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Detection of Seam Carved Image Based on Additional Seam Carving Behavior
Yongzhen Ke,Qingqing Shan,Fan Qin,Weidong Min,Jing Guo 보안공학연구지원센터 2016 International Journal of Signal Processing, Image Vol.9 No.2
Seam carving is a kind of content aware image retargeting algorithm and can be applied to resize and deliberately remove objects from digital images. Based on the observation that after applying an additional seam carving operation, the similarity, the energy relative error, and the difference of seam distance of original image are quite different from those of the seam-carved image, we propose and develop a new method for detecting seam carving or seam insertion of natural images without knowledge of the original image. First, we apply an additional seam carving operation to the testing image, then calculate similarity, energy relative error, and difference of seam distance between the testing image and its seam carved version. Last, we extract 11 dimensional features to detect seam carving operation to train a support vector machine classifier for recognizing whether an image is an original or it has been modified using seam-carving. Our experimental results demonstrate that our proposed forensic method achieves not only better detection rate but also lower dimensional features compared with other existing seam carved detection methods.
An Efficient Blind Detection Algorithm of Median Filtered Image
Yongzhen Ke,Fan Qin,Weidong Min,Qiang Zhang 보안공학연구지원센터 2015 International Journal of Hybrid Information Techno Vol.8 No.1
Due to the significant advances of digital photography and the availability of many powerful photo editing tools, it becomes easier to create forgery images by non-professional users. Median filtering that is usually applied to erase the forensically significant fingerprints has recently received increased attention. In this paper, we present an effective blind forensic algorithm to detect the median filtering manipulation. First, the median filtered residual (MFR) is generated by computing the difference between a testing image and a median filtered version of itself. Then, three feature sets including histogram, autocorrelation and gradient are extracted from the median filter residual. Last, those features are fetched into support vector machine (SVM) for training and classification. Our experimental results demonstrate that our proposed forensic method achieves not only better detection rate but also lower computational complexity compared with other existing median filtering detection methods. Our proposed forensic method also can locate local median filtering of image effectively.
Human Fall Detection Based on Motion Tracking and Shape Aspect Ratio
Weidong Min,Longshu Wei,Qing Han,Yongzhen Ke 보안공학연구지원센터 2016 International Journal of Multimedia and Ubiquitous Vol.11 No.10
Automatic human fall detection based on video monitoring plays an important role in protecting vulnerable people especially the seniors whose falls could cause severe injuries and need attentions from others immediately. In this paper, an automatic human fall detection method based on human motion tracking and shape aspect ratio in real-time video is proposed. While most existing methods detect falls in homes, the proposed method is suitable for both outdoor and indoor environments. The method first detects human objects in general environment, then tracks their motions, meanwhile calculating and recording the motion characteristics of each person. Comparing with the existing fall detection method using shape aspect ratio, the proposed method has advantages of employing the shape aspect ratio together with the moving speed and direction to better detect human falls, as well as being able to detect falls toward different directions. Experiment results demonstrate that the proposed method can effectively detect human falls in general environments including outdoor places.