Browsing by Author "Yin, Jin-Sheng"

Now showing 1 - 7 of 7

A visible wavelet watermarking technique based on exploiting the contrast sensitivity function and noise reduction of human vision system
(2014-09-01)
Tsai, Min-Jen
;
Liu, Jung
;
Yin, Jin-Sheng
;
Yuadi, Imam
;
資訊管理與財務金融系註：原資管所+財金所
With the widespread use of the Internet and the rapid development of digital technologies, copyright protection of multimedia content has become an important issue. Among the available technologies, digital watermarking techniques are regarded as a solution to the property right protection for multimedia resources. To evaluate the performance of a visible watermarking technique, robustness and perceptual translucence are two essential criteria for the watermark applications. In order to get the best trade-off between the embedding energy of a watermark and perceptual translucence, this study presents a technique named ICOCOA (innovated content and contrast aware) by exploiting the contrast sensitivity function (CSF) and noise reduction of human vision system in the wavelet domain. Another novel idea of this work is to propose the innovated CSF masking (I-CSF) curve which provides better weight perception where a game-theoretic architecture can be leveraged to determine the best I-CSF masking for the watermarked image. The experimental results demonstrate that the proposed approach not only provides a good translucent quality of the watermark but also achieves the robustness against the common image processing operations.
Digital Forensics for Printed Character Source Identification
(2016)
Tsai, Min-Jen
;
Hsu, Chien-Lun
;
Yin, Jin-Sheng
;
Yuadi, Imam
;
資訊管理與財務金融系註：原資管所+財金所
Even digital content is widely used nowadays, printed documents are still ubiquitously accepted and circulated. Therefore, identifying the printed character source is essential for criminal investigations to authenticate the digital copies of the printed documents. This study carefully examines the important statistical features from Gray Level Co-occurrence Matrix (GLCM), Discrete Wavelet Transform (DWT), Spatial filters, Wiener filter, and Gabor filter to identify the printer source for Chinese characters by using support vector machine (SVM) and decision fusion of feature selection. Even the subject of printed Chinese character source identification has been investigated, the proposed technique further expands the feature space which achieves superior experimental results and outperforms the techniques described in the literatures. Therefore, the methodology proposed in this study can accomplish high classification accuracy rate which show promising applications for real world digital forensics.
Digital Forensics of Printed Source Identification for Chinese Characters
(2014-01-01)
Tsai, Min-Jen
;
Liu, Jung
;
Yin, Jin-Sheng
;
Yuadi, Imam
;
資訊管理與財務金融系註：原資管所+財金所
Recently, digital forensics, which involves the collection and analysis of the origin digital device, has become an important issue. Digital content can play a crucial role in identifying the source device, such as serve as evidence in court. To achieve this goal, we use different texture feature extraction methods such as gray-level co-occurrence matrix (GLCM) and discrete wavelet transform (DWT), to analyze the Chinese printed source in order to find the impact of different output devices. Furthermore, we also explore the optimum feature subset by using feature selection techniques and using support vector machine (SVM) to identify the source model of the documents. The average experimental results attain a 98.64 % identification rate which is significantly superior to the existing known method by 1.27 %. The superior testing performance demonstrates that the proposed identification method is very useful for source laser printer identification.
Japanese Character based Printed Source Identification
(2015)
Tsai, Min-Jen
;
Hsu, Chien-Lun
;
Yin, Jin-Sheng
;
Yuadi, Imam
;
資訊管理與財務金融系註：原資管所+財金所
Due to the advances of digital technology, digital contents are widely accepted all over the world. Since the criminal counterfeits and copyright infringement is still under study for the new digital media, there is an increasing need called digital forensics, which involves the collection and analysis of the original digital device, has become an important issue. This study offers a universal approach to extract the important statistical features from Gray Level Co-occurrence Matrix (GLCM), Discrete Wavelet Transform (DWT), Spatial filter, Wiener filter and Gabor filter to identify the printer source for Japanese characters. The average experimental results achieves high identification rate at 94.23% which is significantly superior to the existing known method of using GLCM at 77.31%. The result is also better than the technique for Chinese character identification method at 88.77%. This superior testing performance demonstrates that the proposed identification technique is very effective for Japanese character based source laser printer identification.
Source Identification for Printed Documents
(2017-01-01)
Tsai, Min-Jen
;
Yuadi, Imam
;
Yin, Jin-Sheng
;
Tao, Yu-Han
;
資訊管理與財務金融系註：原資管所+財金所
Technological advances in digitization with a variety of image manipulation techniques enable the creation of printed documents illegally. Correspondingly, many researchers conduct studies in determining whether the document printed counterfeit or original. This study examines the several statistical feature sets from Gray Level Co-occurrence Matrix (GLCM), Discrete Wavelet Transform (DWT), Spatial filters, Wiener filter, Gabor filter, Haralick and fractal filters to identify text and image document by using support vector machine (SVM) and decision fusion of feature selection. The average experimental results achieves that the image document is higher identification rate than text document. In summary, the proposed method outperforms the previous researches and it is a promising technique that can be implemented in real forensics for printed documents.
Tree group based Wavelet Watermarking using Energy Modulation and Consistency Check (WW-EMCC) for digital images
(2015-12-01)
Tsai, Min-Jen
;
Yin, Jin-Sheng
;
Yuadi, Imam
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Wavelet tree based watermarking algorithms are generally using the energy difference among grouped wavelet coefficients for invisible watermark embedding and extraction. According to cryptanalysis of wavelet tree quantization (WTQ) scheme, the robustness of watermarking is weak if the wavelet tree group coefficients are only unilaterally modulated. Therefore, bilaterally modulated techniques like modified wavelet tree quantization (MWTQ) and wavelet tree group modulation (WTGM) improve the security since the attackers can not decipher how tree coefficients are modulated. However, MWTQ needs the wavelet tree group information as the extra information which results the method is not purely blind for watermark extraction. For that matter, a novel wavelet tree group based watermarking using energy modulation and consistency check (WW-EMCC) is proposed in this study which not only resists the cryptanalysis attacks but also provides the dual function of choices for blind (WW-EMCCB) and non-blind (WW-EMCCN) watermark embedding. The essence of WW-EMCC design is to embed the watermark in the tree group coefficients as well as the relationship between the tree groups. Such approach extends the bilateral modulation into higher dimension of modulation and increase the robustness of security. In addition, WW-EMCC can even be modified as a captioning watermarking with lossless image quality which integrates watermarking and cryptography for copyright protection. This study has performed intensive comparison for the proposed scheme with WTQ, MWTQ and WTGM under various geometric and nongeometric attacks. The experimental results demonstrate that the proposed technique yields better performance with higher degree of robustness.
Using decision fusion of feature selection in digital forensics for camera source model identification
(2012-03-01)
Tsai, Min-Jen
;
Wang, Chen-Sheng
;
Liu, Jung
;
Yin, Jin-Sheng
;
資訊管理與財務金融系註：原資管所+財金所
Digital forensics, which identifies the characteristics and origin of a digital device, has become a new field of research. If digital content will serve as evidence in court, similar to its non-digital counterparts, digital forensics can play a crucial role in identifying the source model or device. To achieve this goal, the relationship between an image and its camera model will be explored. Various image-related and hardware-related features are utilized in the proposed model by a support vector machine approach along with decision fusion techniques. Furthermore, the optimum feature subset to achieve the highest accuracy rate is also explored. (C) 2011 Elsevier B.V. All rights reserved.