Tsutomu Horikoshi

Feature extraction of temporal texture based on spatiotemporal motion trajectory

A framework and method are proposed to extract local features of a certain kind of naturally occurring, non-rigid motion pattern, referred to as temporal texture. To catch both the spatial and temporal features of this complex pattern, we focus on the surfaces of motion trajectories in spatiotemporal space derived from multiple frames of an image sequence, and represent the surfaces as a set of tangent planes of the surfaces. From the distribution of the tangent planes in local regions in time and space, spatial and temporal texture features are computed The features considered here include spatial arrangement of dominant contours, uniformity of velocity components, and trajectory run length. Experimental results show that the newly defined features have the capability of quantifying the features of complex motion patterns such as weather radar images.

Image velocity estimation from trajectory surface in spatiotemporal space

A new framework and method, based on image motion trajectories in spatiotemporal space (x-y-t space), are proposed to estimate image velocity from an image sequence. We focus on the surfaces of the trajectories in the x-y-t space formed by the edges and contours of moving objects and obtain image velocity from the orientation of the intersection line formed by tangent planes on the trajectories. The proposed method includes two Hough transforms to detect the most dominant orientation in all possible intersection lines and reliably produces the dominant translational image velocity semi-locally. Also, the confidence measure of estimates is defined to decide the optimal size of patch that suppresses the aperture problem. Experimental results from several synthetic and real image sequences are presented to verify the effectiveness of the method and to confirm its robustness against noise and occlusion.

Time-sharing display approach using liquid crystal light valve and a photoreractive crystal for electroholography

A new electroholography system is introduced that uses a liquid crystal display, a photorefractive crystal, and a time- sharing display method. By using the photorefractive crystal as a kind of screen through which the hologram is recreated, we can realize high quality, off-axis, and computer-generated holograms. The time sharing approach enables an electronic display device to display 3D objects clearly by breaking them into parts that are displayed in sequential frames. If the fames are cycled at a high enough speed, we see these parts as a single object due to the afterimage effect. This approach relaxes the strict dynamic range requirements that would normally be placed on the electronic display device. The photorefractive crystal suppresses the flicker caused by the time-sharing display method. This paper first describes the time-sharing method. Next, a new experimental 3D-TV system with a photorefractive crystal, strontium barium niobate, is demonstrated. This system confirms the exciting future of electroholography.

Memory-based forecasting of complex natural patterns by retrieving similar image sequences

A novel framework called memory-based forecasting is proposed to forecast complex and time-varying natural patterns. In this framework, past patterns similar to the present pattern are retrieved, and the forecast pattern is produced by using the patterns that follow the retrieved sequences. We represent the dynamic features of a sequence by using the spatial distribution of the patterns, velocity field, and temporal texture features; the sequences are transformed into paths in eigenspaces. The relationship between retrieval error and prediction error is modeled to define a similarity measure for retrieval. Forecast images are constructed from a future point in the eigenspace which is estimated by a nonlinear prediction scheme. Several experiments using weather radar images confirm the effectiveness of our method especially for drastically changing patterns.

Computer-generated holography using complex interference patterns

This paper introduces a scheme whereby complex holography is realized as computer generated holograms. Complex holography is based on the complex amplitude of the wavefront using an additive representation of a real part and imaginary part. Two types of hologram are needed, but no conjugate image is produced. Conventional holography can avoid the effects of conjugate images only by using a display/recording device with very high resolution, but this solution is not available to electro-holography. The conjugate image, which is always generated from the original image simultaneously, makes the reconstructed image very noisy. Complex holography involves decomposing the complex value of each pixel into the real and imaginary parts of the pattern. Each interference pattern is displayed on a different display device, and illuminated by coherent lights. At this time, the coherent lights differ only in terms of phase, the phase difference is (pi) /2 radians. The summation of the two diffraction patterns regenerates the complex wavefront while eliminating the conjugate object light wave-front. This paper first mathematically confirms the principle of the proposed method. Next, experimental simulations in which synthesized images are clearly reconstructed are described. In addition, an experimental system is demonstrated that uses two LCD panels.

Feature Extraction and Classification of Cell Morphology in Response to the Cell Cycle

The cancer cell exhibits a cyclic change in its morphology, and the discussion of the cyclic change in cell morphology is very important in cytology, considering the relation between the cell growth and the distribution of the protein granules. At present, this kind of analysis is made by human observation through a microscope. This paper describes a method which recognizes the cyclic morphological changes of the cancer cell by image processing. First, various feature parameters of the nucleus and the cytoplasm are calculated by computerized image processing. In the extraction of the cell image from the original image, the binarization rather than ternarization is applied separately to the nucleus and the cytoplasm. Based on the calculated feature parameters, three decision criteria are set based on the discriminant analysis to recognize the cell morphology. As a result of experiment, the recognition rate of 93.9 percent was achieved for the training data, indicating that the cell morphology can be recognized by a computer.

Multiscale structure from multiviews by ⊇2G filtered three-dimensional voting

A hierarchical matching is the most efficient way to recognize a 3-D object. In most conventional methods of 3-D object recognition, a hierarchical structure is formed after the detailed 3D structure of an object has been extracted by dividing it coarsely. This approach is very inefficient, since it is difficult to obtain a detailed 3D structure initially in addition to a large amount of computation time. This paper proposes a method which extracts a multiscale 3D structure directly from coarse to fine. In this method, a space containing an object is divided into coarse voxels, and a “3D voting” is applied. This voting adds a certain value to a voxel through which a back-projected line connects the center of the camera lens and a feature point on an image. A voxel having a high value is regarded as a region containing a 3D feature point, and this is extracted. Extraction of a hierarchical structure is carried out by repeating the dividing process from coarse voxels to fine voxels so that the operation becomes efficient. This paper also analyzes the effects of the deviation of a feature point in an image and the deviation of the camera center on the 3D voting. The paper shows also that an application of a ▽2G filter to the voxel space (after 3D voting) is effective in suppressing the errors due to the deviations. The method has been confirmed successfully by processing six complex test images.

Superquadrics for 3‐D shape indexing language

With the recent development of input, presentation, and processing technologies for three-dimensional (3-D) images as the background, the demand for the effective storage and management of 3-D images is expected to increase. This paper aims at the effective storage and retrieval of the 3-D images and proposes a 3-D shape indexing by the superquadrics. The features of super-quadrics are that the expression is simple and various shapes can be represented using a smaller number of parameters. To examine the applicability of the function, the technique based on superquadrics is compared with conventional 3-D modeling from the viewpoint of the power for representing the feature and shape as the technical requirement for the indexing. The purpose of the conventional modeling technique is to represent the shape from the visual sensation. However, this resulted in a problem whereby it is difficult to determine the shape from the modeled data structure. By contrast, with superquadrics, the functional parameters are easily related to the shape; thus realizing a sufficient applicability as the indexing. An image retrieval system is constructed using the superquadrics parameters as the indexes, and the usefulness of the proposed indexing technique is demonstrated.

Image processing algorithm for edge‐line extraction of lymph vessel wall and measurement of the vessel diameter

The mechanism of the lymph flow is not fully understood. Since the lymphatic system has no pump apparatus such as a heart in the blood vessel system, it is important to study the relationship between the lymph flow and the contractile movement of its wall. This paper analyzes the rhythmical contraction of the lymphatics of the rat mesentery in vivo. Only the image of the lymph vessel wall was extracted from the microscopic image of the mesentery recorded on video tape. An algorithm was developed which was suited to the continuous measurement of the vessel diameter. In the process of the automatic extraction of the lymph vessel wall, not only its derivative and the direction, but also the running state of the lymph vessel were utilized as information. A circular scanning was employed for the first time as a method which realized an accurate diameter measurement of the vessel that always changed its shape with time. By the proposed image processing system, the periodic movement of the lymph vessel could be analyzed quantitatively.