Subjects were placed, in complete darkness, and asked to elicit a saccade from a repeatedly flashing stimulus in the center of the visual field to a saccade target located 10 deg to the right of the stimulus. The stimulus produced several phantom flashes basically in the direction of the saccade; they were mislocalized to the saccade target, when the flashing frequency was high (e.g., 500Hz). The stimulus was extinguished near the saccade in time. The subjects reported the position of the finally perceived phantom flash. The results of the two experiments showed that the degree of mislocalization decreasing flashing frequency, but it is not affected by the luminance of each flash or the average luminance of the flashing stimulus. Our previous study suggested that the degree of mislocalization depends on the attended position during the saccade. The results of the present study were discussed related to the attention shift.
Generally, decreasing the depression angle of HUD images, drivers can perceive more information displayed on the HUDs. However, when distance to the displayed image is not so long and the depression angle is very small, drivers sometimes perceive “double-images” displayed on the HUDs. We have first proposed and verified the hypothesis that the “double-images” stems from driver's convergence response delay.Observing driver's eye movement using EOG, we have found that the convergence response closely depends on the depression angle of HUD image. We have also found a finite depression angle, causing version of the driver's eyes to watch the HUD image, shortens the rise-up time of the convergence response and then improves the “double-images”. Next we examined how the version works to improve the “double-images”, focusing on a saccadic suppression of visual information. Finally, we have developed a cognitive model that can explain the process of representing objects as HUD images at varying distances within a 3D space.
A simple reaction time for the auditory modality is significantly shorter than that for visual modality. When stimuli in both modalities are presented, however, human beings often give preference to the visually presented information. This cannot be explained by the reaction time for each modality. The present experiment has examined whether visual stimuli were preferred or not in a spatial-choice task that requires a saccadic eye movement to respond. Light and sound stimuli appeared at different positions almost simultaneously, so subjects had to move their eyes quickly to one side. The results showed visual preference in this spatial-choice task, and this effect was stronger than in spatial-choice tasks that require responding by pushing buttons. We have discussed the relations between these phenomena and orienting attention model.
A tumor stimulates the growth of small blood vessels for feeding itself. Tumor-derived small blood vessels are an intrinsic part of the tumor's development and progression. Radiographic depiction of tumor-derived small vessels is a useful tool for a basic medical study of malignant tumors. A high-resolution X-ray imaging system with a spatial resolution in the micrometer range was developed at SPring-8 for the depiction of newly formed blood vessels in animal models of cancer. The imaging system is composed of two types of detectors. One is an X-ray indirect-conversion type detector, which is a fluorescent-screen optical-lens coupling system using a high-sensitivity HARPICON pickup tube. The other type is an X-ray direct-conversion type detector, which incorporates an X-ray direct-sensing SATICON pickup tube. Small tumor blood vessels with diameters of around 15μm were observed after injecting a contrast material to the artery in experiments using small animals.
Elderly people tend to perceive color as more yellowish than younger people do, and their ability to discriminate between blueish colors deteriorates as they grow older. This paper describes measurements of visual chromatic spatial and temporal frequency responses in subjects aged 60 to 75 years old to gain a better understanding of age-related changes in color perception and to clarify some differences between young and elderly people. We found that visual chromatic spatial and temporal frequency responses to ayellow-blue (dominant wavelength : 565-445 nm) stimulus were significantly reduced in comparison with those to a red-green (495 nm and its conjugate) stimulus. The results suggest that weakened optical functions in the eye and in the power to send visual signals from the retina to the nervous system are likely causes of this deterioration of color -contrast sensitivity.
We investigated the effect of frame delay on stereoscopic vision by using a moving object in a virtual environment. The object was translated along the x-, y-, or z-axis or was rotated around the x-, y-, or z-axis while changing its speed and size. The effect on stereoscopic vision was measured by using the method of limits by changing the delay (i. e., the number of frames) between left-eye image and right-eye image. The measurements clarify the change tendencies with a maximum number of delays and allow the viewer to enjoy clear stereoscopic images when frame delay occurs.
A speech recognition system with a manual error correction has been developed to produce closed captions in live broadcasting programs such as news programs. Speech recognition errors, however, are not corrected completely because the correctors often miss errors due to lack of care or successive erroneous words. In this paper, we propose a method that detects errors automatically to assist manual correction. A coustic parameters were extracted from both correct and erroneous results of all morphemes produced by speech recognition systems. Templates which can precisely distinguish between errors and correct results, were then constructed by genetic algorithms (GA) and discriminative training. Consequently, theerrors are detected by comparing the acoustic parameters of an unknown recognition result with the templates. Experiments have confirmed that presenting erroneous words by using the proposed method is effective for improving corrector error detection.
A 3D stereo image makes a more powerful and realistic impression than a 2D image. We want to clarify the optimal projection conditions for educational facilities and amusement arcades that use 3D stereo images. First, we clarified the degree of impression due to screen for 2D and 3D stereo images. The impressions of 3D stereo images became more powerful and realistic than those of 2D images as the screen size was widened. We statistically examined the opinions of viewers of 2D and 3D stereo images, which were obtained from questionnaires. Visual fatigue sometimes appeared after they watched 3D stereo images, specially when a projection method using two projectors (one for the right-eye and the other for the left-eye) was used. This fatigue appears to be partially related to the projection-size difference. Visual fatigue became more serious when the difference extended beyond 1% of the original size. Using these results, we designed an assembly rack for two projectors (3D stereo projectors) to keep the size difference below 1%.
Creating reactive motions with existing motion-capture systems is difficult because they require a variety of task environments. To overcome this drawback, we propose a reactive motion capture system that combines the existing motion capture system with a haptic device and a human-scale virtual environment. Our objective is to acquire reactive motion data generated from the interaction between haptic display units and the virtual environment, using the fact that a person's actions in actual situations can be represented by the reactions of the person to actual objects. We asked the subject to trace the rim of the front of a cube under six task conditions and measured their finger trajectories as motion data. The results demonstrated that our proposed system is useful for generating data for natural-looking motions similar to real-world motions.
Adaptation speed improvement in adaptive equalizer is proposed and evaluated for relay stations in the Japanese digital terrestrial broadcasting system, ISDB-T (Integrated Services Digital Broadcasting for Terrestrial). Adaptive equalizers are advantageous in hardware implementation, but their cancelation performances for time variable environments have not been studied. In this paper, the use of tap selective approaches are proposed for LMS and RLS algorithms. Computer simulations show that the methods for achieving high performance are applicable to actual relay station systems in time variant transmission environments.
We developed a high efficiency power amplifier with a pre-distortion circuit that uses temperature compensation in the ISDB-T STL/TTL in the SHF band. This equipment works in the 7GHz band and greatly reduces distortions caused by non-linear characteristics of the amplifier by using a pre-distortion circuit. The amplifier consists of a pre-distorter, a power amplifier section, a control circuit for the temperature compensation and an ALC circuit to keep the output power constant. It maintains a great stability during temperature changes by controlling the input level of the pre-distorter according to the measured temperature in the power amplifier section. This amplifier has an excellent C/N ratio of 53 dB with an output of 2 W (OFDM average power) and a power efficiency of 2.2%.