Which lighting attributes determine the brightness of a room space? We considered the perceived strength of the light source and the perceived amount of light in the space as potential determinants. We examined the relations between the actual brightness of the space and these two perceptions. In the first experiment, subjects evaluated the strength of the light source and the amount of light in the space using reference matching method. The results were consistent among the subjects, indicating they judged each of the two lighting attributes by some common criteria. In the second experiment, the brightness of the space, the strength of the light source, and the amount of light in the space of a model box were evaluated using the magnitude estimation method under various illumination conditions. We found that the brightness of the space was closely correlated with the perceived amount of light but not with the perceived strength of the light source. These results should prove useful in constructing a model for quantitatively estimating the brightness of a space.
The operating life of a fluorescent lamp is determined by the life of its cathode, which depends on how quickly the cathode loses its oxide coating. There are two mechanisms, vaporizing and sputtering, in this process. The sputtering is affected by the current crest factor (CCF) because the electron current is determined by the cathode temperature and the electric field of the cathode fall: Ie a T2exp (-e/kT) exp (0.4E/T). When CCF is high, the cathode temperature cannot follow variations in the current because of thermal inertia. As a result, the cathode fall voltage (CFV) increases in order to supply the flow of peak electron current, which leads to heavy sputtering. The equation for electron current also shows that CFV can be reduced by heating the cathode auxiliarily if CCF is high. Examination of the relations among CCF, CFV, and the cathode temperature measured by the indirect method showed that (1) The peak CFV increases as the CCF becomes higher when the cathode is not heated. (2) The peak CFV can be reduced by heating the cathode if the CFV is high. (3) The heating power needed to reduce the CFV is small enough for the practical circuit.
A method is described for detecting a light-source position that can be used to find partial discharges in gas-insulated switchgears as an indication of a fault. Comparing the illuminances measured by light sensors to those in a database gives information about the light-source position. The database is built beforehand using illuminance distributions, computed by the Monte Carlo method, for each given light source in a lighting space. In the database, the set of measured illuminances for a light source is treated as a position in a fictitious space or plane. This method was tested for a simple cubical room by using the Monte Carlo method to compute illuminances created on sensors by test light sources and by estimating the accuracies of detection. The maximum effective accuracy obtained when expanding the detected space to a neighborhood space was 86.9%. The detected accuracy improved that the light sensors were placed near a corner of the room. The positional arrangement of the light sensors, therefore, should be optimized. Building the database more or less affected the accuracy as well. Once the database is built, the calculation should take little time. This method should thus be useful for quickly identifying fault points in gas-insulated switchgears.
A method is described for reducing the large error that often occurs in the measurement of three-dimensional (3D) objects using the photometric stereo method when there is a specular component in the reflection characteristics of the object's surface. First, the change in the luminance distribution caused by the component is estimated based on the correlation between the modeled distribution and a gray-level image of the object, created by numerical experiment. Next, the shape measurement of the hemisphere object was carried out. Numerical experiment of a hemispherical object showed that the mean error of normal vector can be reduced by as much as 90%. Thismethod is applicable to machine vision, automatic manufacturing, and other industrial processes.
The cell density of Pavlova lutheri was measured during a 17 days-long culture period at the three different-sized culture vessels; a 3l flask, a 20l tank, and a 100l tank. PPFD (Photosynthetical Photon Flux Density) at the center of the 3l flask and the 20l tank under fluoresent lamps was 45μmol· m-2·s-1, while that at the 100l tank was 300μmol·m-2·s-1 under high pressure sodium lamps, and 0-1300μmol·m-2·s-1 under the natural sun light. The cell density was 2.1×107 cells·ml-1 at the 3l flask, 7.4×107 cells·ml-1 at the 20l tank, and 1.6×107 cells·ml-1 at the 100l tank. Differences at the cell density were found in between the 3l flask and the 20l tank, though the same light sources were used for these two systems. The cell densities at the 100ltank were also different when high pressure sodium lamps and natural sun light were used, however no difference at the cell density was found between the 100l tank under high pressure sodium lamps and the 3l flask under fluorescent lamps. Through these results, it can be concluded that a propagation rate of P. lutheri increases under higher PPFD. For higher scaling up of vessels will increase under higher PPFD lighting. It is necessary to examine the much more efficient lighting in order to maximize a propagation rate of the micro algae.
At dusk, the change in brightness of peripheral vision is striking. Such a sudden change in the viewing environment could affect the vision of a driver, possibly leading to an accident. I investigated the occurrence of traffic accidents at dusk by statistical analysis to determine whether changes in the viewing environment at dusk affect the accident rate. Dusk is considered to be the time of civil twilight, which is about 30 minutes in Tokyo. The analysis was done under certain conditions; for example, the target area was limited so that the time period for dusk was basically the same. Furthermore, the effect of the day of the week on the occurrence rate was investigated. I found no tendency for traffic accidents to occur with higher frequency at dusk. Moreover, the analysis of the accident rate of the day of the week showed that the accident rate is rather related to the traffic volume.
In an inductively coupled electrodeless metal halide lamp dosed with NaI and ScI3, dosing a small amount of SnI2 into its arc tube suppresses the free I2 gas generation, which extends the life of the lamp. The free I2 gas is thought to be left over as a result of Na and Sc loss reaction with the quartz. It is assumed the Na loss is caused by its migration into the quartz and that the Sc loss is caused by the reaction between the quartz and Sc clusters on the quartz. The Na loss rate depends on the partial pressure of the Na and the Sc loss rate depends on the nucleation rate of the Sc clusters, which can be derived from the nucleation theory of clusters. Thermodynamical calculations showed that the Na and Sc partial pressures become 1/100-1/10000, and the nucleation rate of the Sc clusters becomes negligible with SnI2 dosing. The rate of the free I2 gas generation was calculated, assuming it is proportional to the Na pressure and the Sc cluster nucleation rate. The proportional coefficients were determined from the results of lifetime tests and furnace-heating tests. The calculations showed that the SnI2 dose sharply reduces the rate of free I2 gas generation, making the lamp-life longer.
Although it is highly desirable to be able to examine the contrast of luminance quantitatively in order to design an appropriate brightness pattern, conventionaluminance contrast expressing methods can not be applied to the actual complicated luminance distribution. A previously proposed analysis method for expressing the luminance distribution was used to develop a technique for quantifying the contrast of luminance in an actual complicated luminance distribution. Application of this technique showed that it works well to actual distributions.