The style of cryptography, in which important data are embedded into an image not to be noticed by outsiders, is called Video Steganography. In Video Steganography, if a lot of data are embedded into an image, the image, the image will be distorted. If such a distorted image is transmitted, outsiders may suspect that the image is a cryptogram. In this report, we propose a new method of data embedding for Video Steganography. In thid method the amount of locally embedded data is adaptively varied according to the local characteristics of the original image, and more data can be embedded into the image with less distortion. Uing the discrete cosine transform (DCT), data are mainly embedded into high frequency components of the image, which have few influence on the quality of the image. The amount of embedded data is varied depending on DCT components of adjacent areas not to weaken edge lines in the image. Finally, we show the effectiveness of our method by subjective evaluation.

View full abstract

A cryptogram embedded in a image data which has other meaning is calle video steganog raphy. This paper describes a data embedding technique for bilevel images using the error diffusion method and the mean density approximation method. In these methods, the threshold for a current point is determined by values of its neighborhood. Therefore, if data embedding and reducing to bilevel are done simultaneously, noise by embedded data would be recovered by thresholding. This technique can also apply to multilevel images.

View full abstract

Hilbert curve is one of Peano's space-filling curve published in 1890. There are several applications of this curve such as image processing, compiter hologram, etc. In this paper, we concentrate on a lossy compression for a gray image using the Hilbert curve. The merit of this scan is to pass through all points in a quadrant and always move to the neighbor quadrant. Our method is a compression of a gray image by a simple segmentation of this scanned signal using a zero order interpolation. From our experiments, we have confirmed that inspite of the simple computation in comparison to JPEG, the reconstracted image which compression rate is about 0.6blt/pixel has acceptable quality.

View full abstract

There are several algorithms for N-dimensional Hilbert scanning, such as the Butz algorithm and the Quinqueton algorithm. The Butz algorithm is a mapping function using several bit operations such as shifting, exclusive OR, etc. On the other hand, the Quinqueton algorithm computes all addresses of this curve using recursive functions, but takes time to compute a one-to-one mapping correspondence. Both algorithms are complex to compute and both are difficult to implement in hardware. In this paper, we propose a new, simple, non-recursive algorithm for N-dimensional Hilbert scanning using lookup tables. The merit of our algorithm is that the computation is fast and the hardware implementation is much easier than previous ones.

View full abstract

We propose a production method to generate and control computer graphics animation using a giant virtual spreadsheet called Animation Spreadsheet (AS). AS contains thousands of images as miniature image data which can be designed, generated, edited and saved. AS has two axes. One is based on time and the other is on layers of image components which will consist a cell of completed image. Users are able to manipulate images by previewing with time axis or with layer axis. AS can be used not only for computer generated animation production but also as designing and editing tool for other video image production. We examined AS on a short animation production to compare with existing normal production method.

View full abstract

In conventional studies on ray tracing techniques in computer graphics. there is an assumption that light beams go straight on. But in layers of different density matters or in a gravitational field, light beams don't go straight on. This paper concerns a ray tracing method in a gravitational field which meets the theory of relativity. We generate a strong gravitational field by putting a black hole in a vacuum space. And we study on the visualization of moving objects near the field. In this paper, we lead differential equations of ray and object on Riemann manifold from a model of space which has a black hole at the origin, then solve these equations using the variable step-size numerical method, and finally generate images of a moving object in the gravitational field.

View full abstract

A complicated surface modeling with volume representation usually requires a lot of memory. On the other hand, repeatedly re-using a few kinds of volume data(voxel) for saving memory often causes unnatural regular patterns. In this paper, to avoid this difficulty, we propose a method of volume modeling and rendering with classified density distribution. A lower class of density distribution represents a primitive shape of an object. A higher class determines a geometric layout of the lower one. We also show some examples of rendered images such as a forest scenery.

View full abstract

High speed calculation for Fresnel Hologram is very important to the continuous real-time display of a holographic 3D image. We are studing computer system and algorisms for Computer Generated Hologram (CGH). The Fresnel Hologram was calculated using a hardware accelerator. Calculation speed was thirty times faster than the conventional workstation. We proposed higher speed and more accurate new Fresnel Hologram calculating system for CGH.

View full abstract

A new 3D image display using multiple liquid-crystal panels is proposed, that gives true 3D images with wide viewing angle. Prototype system consists of 9 liquid-crystal panels (213mm (H) ×133mm (V)) aligned at intervals of 15mm depth, that can display about 10cm^3 object. Moving 3D images can be realized at a frame per from 0.7 to 1.5 sec. by producing and storing sliced images generated from original voxel data. The resulting image can be observed from any distance and at wide viewing angle (≨21゜).

View full abstract

In this paper, we report an experimental evaluation of depth representation in a 3-D visualization. To investigate depth representations in real-time 3-D interactive graphics. two experiments are performed : Experiment 1 is to evaluate effectiveness of motion parallax and binocular stereopsis with node-searching tasks on 3-D directory trees (Cone Tree) ; Experiment 2 is for examinig the effectiveness of a head coupled display system with Cone Tree and Spiral Tree. The results show qualitative characteristics of depth representations in a 3-D visualization.

View full abstract

Automatic lipreading requires an automatic lip contour extraction tool. In this paper. based on an active contour model, a lip contour extraction method is presented, which can automatically extract the lip contour from TV face images. It needs neither a special artificial condition i.e., smearing the lips with a black lipstick, nor initializing the start points by human assistance. It accepts one person's TV shot images as its input, and extracts the person's lip contour fully automatically. Experimental results show the effectiveness of the proposed method.

View full abstract

A new method to interpolate correct images at the covered or uncovered background areas is proposed. The covered or uncovered background areas are detected using the motion vectors which are estimated in forward and backward directions. Then the interpolated fields are produced at the covered or uncovered background areas using only one field. In result, the correct images are obtained. It is examined by software simulation that the correct images can be obtained by means of the proposed method.

View full abstract