Jau-Ling Shih

Learn More
In this paper, we will propose an automatic music genre classification approach based on long-term modulation spectral analysis of spectral (OSC and MPEG-7 NASE) as well as cepstral (MFCC) features. Modulation spectral analysis of every feature value will generate a corresponding modulation spectrum and all the modulation spectra can be collected to form a(More)
The advances in 3D data acquisition techniques, graphics hardware, and 3D data modeling and visualizing techniques have led to the proliferation of 3D models. This has made the searching for specific 3D models a vital issue. Techniques for effective and efficient content-based retrieval of 3D models have therefore become an essential research topic. In this(More)
With the increase in the number of trademarks, trademark imitation has become a serious problem. Thus, building an ef®cient trademark retrieval system is imperative. In this paper, such a system is presented. First, a semi-automatic segmentation method is proposed to extract the shapes of those representative objects, called`masks', in each trademark. Next,(More)
In this paper, we proposed a novel feature, called octavebased modulation spectral contrast (OMSC), for music genre classification. OMSC is extracted from long-term modulation spectrum analysis to represent the time-varying behavior of music signals. Experimental results have shown that OMSC outperforms MFCC and OSC. If OMSC is integrated with MFCC and OSC,(More)
In this paper, we will propose a new exterior shape feature, ART-based elevation descriptor (ART-ED), and a new interior shape feature, shell grid descriptor (SGD), for 3D model retrieval. ART-ED describes the elevation information of a 3D model from six different angles. Since ART-ED represents only the exterior contour of a 3D model, SGD is proposed for(More)
As the development of the Internet, children are easily exposed on the pornography through web browsers. To block adult images, content-based image retrieval technique is employed for adult image identification. First, the background is removed to obtain the rectangular region of interesting based on the detection of skin-like pixels. For each input image,(More)
Developing an automatic and intelligent system to detect, track, recognize, and analyze moving objects could save human power in monitoring centers. In this study, the color features of an employee's uniform were extracted to identify the entrance legality in a restricted area of an open space. The body of an entrant was segmented into three parts for(More)