Browsing by Author "Hsieh, Chin-Yu"

Now showing 1 - 2 of 2

A social fundraising mechanism for charity crowdfunding
(2020-02-01)
Li, Yung-Ming
;
Wu, Jhih-Dong
;
Hsieh, Chin-Yu
;
Liou, Jyh-Hwa
;
資訊管理與財務金融系註：原資管所+財金所
In recent years, as the world has witnessed many social issues and environmental disasters, philanthropic giving has become widespread. Today, crowdfunding platforms have also become popular, and charities often use these to fundraise. However, they rarely utilize the power of social networking to assist a fundraiser in discovering potential donors, and this is especially true for small organizations and individuals. In this research, we propose a recommendation mechanism for social fundraising that analyzes the donor's preferences, the relationship between the donor and fundraiser, and the characteristics of the fundraising dynamics to promote the spread of philanthropic fundraising. The proposed mechanism can effectively discover appropriate donors and relevant campaigns to expedite the fundraising process and improve its success rate.
The visual color QR code algorithm (DWT-QR) based on wavelet transform and human vision system
(2019-08-01)
Tsai, Min-Jen
;
Hsieh, Chin-Yu
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
A novel visual Quick Response (QR) code algorithm based on Wavelet transform and Human Visual System (HVS) approach is presented in this study and named DWT-QR. Unlike other QR codes are generally embedded in the spatial domain, the composite coefficients using global and local characteristics of the host image are considered in the discrete wavelet transform (DWT) domain for the visual QR codes. In order to get the best perceptual embedding capability of visual QR codes, the collaboration of the perceptual model of contrast-sensitive function (CSF) with the noise reduction of the visibility thresholds for HVS in DWT domain, achieves the goal of fine tuning of the perceptual weights according to the basis function amplitudes for the best quality of perceptual visibility. In addition, the computation of Noise Visibility Function (NVF) characterizes the local image properties to determine the optimal QR code strength during the QR codes embedding stage. After the detection pattern embedded for the visual QR codes, different distortion attacks have been performed for the proposed method. The experimental results demonstrate that the proposed DWT-QR approach outperforms the known techniques and not only improves the visual quality of the images but also the robustness against various attacks.