In this paper, I propose an extraction algorithm of pitch pulses in voice guidance of car navigation system. The pitch pulses are determined from a pulse like residual signal of the FIR and STREAK filter. (FIR-STREAK filter is combined STREAK filter with FIR filter). As a result, the extraction rate of pitch pulses was 85% for female voice and 96% for male voice. In the Proposed algorithm, The Pitch Pulses are does not average pitch intervals to accomodate the changes in each pitch position resulting from subtle changes in the vocal cords.
Image devices of relatively high resolution are now becoming available. However, for use in pathological diagonosis, high resolution image devices having still more high level of resolutions are strongly required. So we have developed Multi-Computer Multi-Scan Image System which enables to attain the required resolution which exceeds beyond the physical limit of the resolutions of inddividual devices. Based on the idea of Kyoto University, Nikon developed Super High Definition Multi-Image Microscope System composed of two sets of about 2M pixcel HDTV CCD cameras using Optical Multi-Image Device. Kyoto Prefecture University of Medicine advised on all aspects of this system from the standpoint of remote pathological diagnosis, and estimated Super High Definitoin Multi-Image.
Intrasaccadic systematic target step is known to induce adaptation of saccadic amplitude. Former studies of selective adaptation in our laboratory suggested that the adaptation may take place during primary saccades but not in corrective saccades. The present study shows that the adaptation of either visually-guided extemally-triggered saccade or memory-guided saccade have taken place, even when the step back target apperance was consistently delayed by up to 400 msec from the end of a primary saccade. These findings have important implications on the oculomotor learning.
We have shown selective adaptation within two types of volitional saccade pair, and confirmed their adaptation selectivity. Deubel claimed adaptation selectivity only between volitional and reflexive saccades. So we re-examined whether Deubel's scan saccades might have involved two types of volitional saccades. However, our result suggested that scan saccades contain almost one type of volitional saccade, not two of them. Furthermore, our result suggested that scan saccade may be a new type of saccade different from these two types. We may need further experiments.
In the dark, the subjects repeatedly and alternately made saccades between the two horizontally arranged visual targets or their correspondingly remembered targets. Moreover, the intention while the subjects made gaze-shifts between the two points was changed by instruction, i.e., accurately or quickly. The results suggested the following matters. 1. The peak velocity of the saccades elicited toward the visual targets is larger than that toward the remembered targets. 2. The peak velocity of saccade changed when the subjects intended to make gaze-shifts as quickly or accurately as possible. However, the tendency of this change was different between the subjects.
The main purpose of the present study was to investigate adaptive properties in human smooth-pursuit eye movements generated by a peripheral moving target. In adaptation trials, a target appeared in the peripheral visual field and immediately moved away at a constant speed, and a subject made a saccade and postsaccadic pursuit responses to track it. The target speed was. however, changed to a higher or lower constant speed (step-ramp-ramp target motion) at the termination of the initial saccade. This adaptation paradigm induced adaptive modifications in postsaccadic pursuit responses and our results revealed the following properties of the pursuit adaptation system. Topographic modification : Modification of the initial pursuit velocities depends on the position of a moving target. Pursuit gain change : Pursuit velocity is modified not by the addition of a constant bias to the pre-adaptation pursuit velocity, but by a change in the pursuit gain (pursuit velocity/target velocity). Lack of influence on saccade amplitude : Pursuit adaptation did not change the amplitude of saccades either to a moving target or to a stationary target.
Visual illusion seems as one of the reflections of visual mechanisms and it could be an effctive cue for getting better understanding on human visual mechanism. It is expected that the psychophysical studies by using visual illusion can make bridge between psychological researches and physiological researches. Several years ago. the author found new types of visual illusion with binocular viewing, They have been used for investigating the mechanism of human visual system. then several interesting phenomena. have been found. In this paper, the author will introduces newly found 3-D illusions with binocular viewing. Then. he will invesligetes the possibilities of these Illusions to affer one clew for revealing the human visual mechanism.
when the adequate visual stimuli are suitably arranged partially along the surface boundary an opaque or transparent illusory surface canbe seen. Dynamical fusion and separation of illusory surface have been observed when the visual stimuli were canged continuously between two different conditions so that one illusory surface could be perceived in the first conditions and twe separated illusory surface could be perceived in the other condition. Geometrical arrangements for the fusion and the separation have been measured ; then hysteresis of the perception was recognized in the geometorical positions between the fusion and the separation of the surface. In addition, we observed these phenomena with a stationary illusory surface surrounding the dynamically changing illusory surface, then we found that the perception of the fusing and separating positions were chang.
A visual system is able to interpolate occluded edges of objects in a input image even if the edges are curved. It is very important issue to investigate a mechanism that performs the visual interpolation. We propose a visual model for interpolation of occluded curved edges as well as straight edges. The model was applied to gray level images and the results show that occluded edges were successfully interpolated.
Transparency is defined as perception of a surface through another surface as overlapping layers. The physical transmittance of a surface could not be a constraint factor for the transparency perception. In this study. we measured the chromatic and the luminance ranges of the overlapping area of two cross surfaces in order to see transparency. The results show that the additive mixture of colors of the inducing surfaces can be a better constraint factor for transparency. But the luminance of the overlapping area turned out not to be the sum of the inducing surfaces.