Application of Image inpainting ranges from object removal, photo restoration, scratch removal, and so on. In this thesis, we will propose a modified multi-scale method and learning-based method using artificial neural networks for image inpainting. Multi-scale inpainting method combines image segmentation, contour estimation, and exemplar-based inpainting. The main goal of image segmentation is to separate image to several homogeneous regions outside the target region. After image segmentation, we use contour estimation to estimate curves inside the target region to partition the whole image into several different regions. Then we fill those different regions inside the target region separately by exemplar-based inpainting method. The exemplar-based technique fills the target region via the texture synthesis and filling order of exemplary patches. Exemplary patches are found near target region and the filling order is determined by isophote and densities of exemplary patches. Learning-based inpainting is a novel technique. This technique combines machine learning and the concept of filling order. We use artificial neural networks to learn the structure and texture surrounding the target region. After training, we fill the target region according to the filling order. From our simulation results, very good results can be obtained for removing large-size objects by using the proposed multi-scale method, and for removing medium-size objects of gray images.

No

Abstract The original purpose of foundations in Taiwanese society was to remedy the insubstantiality of the government. However, the role of foundations in society has expanded considerably in recent years. The aims of different types of foundations vary a great deal, from focusing on trying to improve basic social welfare, to the support of research investment. In addition, many foundations that originally intervened actively to promote and pursue social welfare have now changed to a far more passive role. This thesis starts by defining the five characteristics of foundations: independence, philanthropy, strategy, transparency, and specialty. It uses these characteristics as a tool to analyze those foundations that were set up by businesses. It is clear from extensive local research that some, if not most, foundations have considerable scope for improvement if they are to become more effective institutions. Taking ¡§independence¡¨ as an example, the purpose of foundations set up by businesses is ostensibly to help people in need, or to help related academic research. However, many business foundations are inseparable from the original mother business. Consequently these foundations have a tendency to lose their original purpose, or become a tool for passing on ¡§favors¡¨. This thesis aims to explain the underlying characteristics of foundations and, by analyzing these characteristics, help to reveal a number of intrinsic problems that need to be addressed.

In multiresolution analysis(MRA)by wavelet function Daubechies (db), we decompose the signal in two parts, the low and high-frequency content,respectively. We remove the high-frequency content and reconstruct a new ¡§de-noise¡¨ signal by using inverse wavelet transform. In order to improve the forecasting accuracy of ANN (Artificial Neural Network) model ,we use the concept of tidal constituent phase-lags, and the new ¡§de-noise¡¨ signal was used as the input data set of ANN. Besides, we also use wavelet spectrum, conventional energy spectrum (Fast Fourier Transform, FFT),and harmonic analysis to analyze the character of tidal data . The results show that the concept of tidal constituent phase-lags can improve ANN model of tidal forecasting and supplement, but in the wavelet analysis , the improvement is insignificant .The reason is that the energy of higher frequency noise is very small compared to the energy of the diurnal and the semi- diurnal tidal components. In other word , the ANN model has a certain tolerance of noise effect .

As the rapid evolution of the technology and the network communication, people can share the multimedia content more easily through the Internet. And devices for accessing the Internet include not only the desktop computers but also cell phones, PDAs, and computers on the vehicles. Vehicular ad-hoc networks, which allow mobile users to access the Internet with the computers on cars is one of the most promising technology in the near future. Through VANET, vehicles can not only share the traffic information to avoid accidents, but also exchange the multimedia content through the peer to peer sharing technology. In this thesis we propose using the network coding technology to encode/decode the file blocks in VANET, clustering the vehicles according to their mobility, and combining the ROMSGP and modifying the priority method of the VANETCODE scheme in order to enhance the overall performance of VANETs.

No description available.

Thesis (M.A.) - Simon Fraser University, 2005. / Theses (School of Communication) / Simon Fraser University. Also issued in digital format and available on the World Wide Web.

Thesis (Ph. D.)--University of Washington, 2002. / Vita. Includes bibliographical references (p. 90-102).

Thesis (M.S. in Computer Science)--Naval Postgraduate School, March 2004. / Thesis advisor(s): Geoffrey Xie. Includes bibliographical references (p. 155-156). Also available online.

Thesis (M. Sc.)--University of Hong Kong, 1964. / Mimeographed.