16-bit PCM will continue to play its role of standard digital storage medium as it did since the advent of Compact Disc, but the entire digital technology movement is directing toward higher-efficiency coding as found in some of recent applications. Defined as emission coder based on TDAC transform. Dolby AC-3 employs a filter bank with critical band principle derived from human auditory model, overlapping windowing for improved block-to-block continuity, and furthermore by processing 5-channel audio data as an entity, its performance is exceptional for a given bit-rate unlike multiplying independent channel coders. This paper also discusses necessary infra-structural establishment for multi-channel programs.
This paper proposes the method of automatical segmentation of oil-painting picture images into the touches of brush using a computer. It is the first step of the research that makes clear the relationship between the features of the touches and the human impression of the picture. First we made the definition of a touch through observation. According to the definition, the image is transformed its color notation, and is performed a region segmentation by two steps of region grawing methods, which are the one using hue and saturation information, and the other using brightness and shape information respectively. It came clear that this method divides a picture image into touches efficiently.
As can be seen in CAD systems, modeling and deforming the shape of 3D objects on computer have been very difficult. In general, the objects are represented as a surface model or a solid model using boundary representation or CSG technique, and the inner structure of the object has been ignored. The deformations are expressed as the relocation of the control points of the parametric surface function. Therefore, it is very difficult, to give the impression and the feeling of the direct manipulation. This paper presents new modeling and deforming technique for interactive tools and proposes a system for designing 3D objects. The algorithm deforms the object by changing the inner structure of the object, not by changing the geometric definition of the surface. And we constructed an experimented system based on this algorithm.
A coreline extraction method based on contour vectors has been proposed. The basic idea of the method is to generate the medial axis from contour vectors sited opposite each other. This method has merits on the time efficiency, the ability to get line width, and the ablity to separate filled area. The method, however, cannot work well on destorted parts. We propose an improved method to overcome the deficiency using the combination method with the coreline extractions and thinning.