Tzu-Wei Huang

Learn More
Automatic photo cropping is an important tool for improving visual quality of digital photos without resorting to tedious manual selection. Traditionally, photo cropping is accomplished by determining the best proposal window through visual quality assessment or saliency detection. In essence, the performance of an image cropper highly depends on the(More)
Since the 1990s, brown root rot caused by Phellinus noxius (Corner) Cunningham has become a major tree disease in Taiwan. This fungal pathogen can infect more than 200 hardwood and softwood tree species, causing gradual to fast decline of the trees. For effective control, we must determine how the pathogen is disseminated and how the new infection center of(More)
We present a novel image representation that characterizes a color image by an intensity image and a small number of color pixels. Our idea is based on solving an inverse problem of colorization: Given a color image, we seek to obtain an intensity image and a small subset of color pixels, which are called landmark pixels, so that the input color image can(More)
Based on wireless local area networks (WLAN), this paper presents the navigation of a robotic wheelchair. We propose a radio based robot localization with a particle filter by using received signal strength from WLAN. An intelligent strategy is proposed to select the informative WLAN signals and reduce the algorithm complexity. Furthermore, we introduce a(More)
This paper presented a brain-computer Chinese Spelling System, which allow the users to spell out Chinese words via his/her brain signals. This system analyzed brain signals by ERPs to find out the subject’s demand. To detect subject’s demand, the flashing phonetic symbols, or Chinese words, were showed in a matrix for subject. The system would totally(More)
We present a new method that addresses the problem of approximate nearest neighbor search via partitioning the feature space. The proposed random exemplar hashing algorithm can be used to generate binary codes of data to facilitate nearest neighbor search within large datasets. Inspired by the idea of using an ensemble of classifiers for discriminative(More)
We present an unsupervised method for discovering objects from depth information. Our method can identify new common objects appearing in different depth images. We use 2D bounding box proposals to detect candidate locations of objects in each depth image, and then retrieve the corresponding 3D bounding boxes using the depth information. Invalid object(More)
Dye-sensitized solar cells (DSSCs) are fabricated with gallium nitride-titanium dioxide (GaN-TiO2) composite photoelectrodes to enhance the power conversion efficiency. The value of power conversion efficiency increases with the incorporation of GaN in TiO2 matrix and reaches a maximum at 0.05 wt% GaN. Internal resistance in the DSSC is characterized by(More)
Yu-Jen Hsiao a, Yen-Hwei Chang a,∗, Yee-Shin Chang b, Te-Hua Fang c, Yin-Lai Chai d, Guo-Ju Chen e, Tzu-Wei Huang a a Department of Materials Science and Engineering, National Cheng Kung University, Tainan 701, Taiwan b Institute of Electro-Optical and Materials Science, National Formosa University, Yunlin 632, Taiwan c Institute of Mechanical and(More)
A pyrene photoluminescent layer is coated on the front side of a dye-sensitized solar cell to enhance the power conversion efficiency. The power conversion efficiency of the dye-sensitized solar cell coated with a pyrene photoluminescent layer improves from 9.12 ± 0.06% to 10.30 ± 0.11% (an improvement of 12.94%).