Online ISSN : 1349-838X
Print ISSN : 0019-2341
ISSN-L : 0019-2341
Volume 101 , Issue 10
Showing 1-2 articles out of 2 articles from the selected issue
Original Papers
  • Tomonori Tashiro, Michico Iwata, Miyoshi Ayama
    2017 Volume 101 Issue 10 Pages 469-473
    Published: October 01, 2017
    Released: December 01, 2017

    Light-emitting diode (LED) light sources are increasingly being used. One LED light source is composed of many small LED chips, so the luminance distribution varies depending on the number and arrangement of the LED chips. Quantitative relationships between non-uniform light sources, such as LEDs, and discomfort glare rating have been widely reported, but the relationship between the non-uniform light sources and the perception of brightness has not been clarified.

    In this paper, a brightness matching experiment was performed to study the effects of non-uniform spatial luminance distribution on perception of brightness. The results show that non-uniform light sources are perceived to be brighter than uniform ones when they have the same average luminance. Brightness sensitivity and discomfort glare rating were also compared. The results show that both the brightness and discomfort glare for non-uniform and/or uniform light sources can be evaluated with the effective glare luminance, which is the index of the discomfort glare rating.

    Download PDF (1682K)
  • Madoka Ohnishi, Koichi Oda
    2017 Volume 101 Issue 10 Pages 474-483
    Published: October 01, 2017
    Released: December 01, 2017

    This study aims to explain as quantitatively as possible the readability of Japanese characters in terms of certain elements in the multi-dimensional, and rather qualitative, space of font design. Based on a review of previous studies, two main dimensions have been selected as the most influential on readability ; the relative character size in the bounding box, which we call style, and the stroke width, which we call weight (when discussed in a categorical way). The Gothic fonts were chosen as the target because they are well known to be the most legible. Behavioral evaluations of readability were conducted instead of subjective judgments. In accordance with the reading acuity measurement MNREAD-J, short and easy-to-read sentences were presented to participants, and the time required to read them aloud was recorded along with any reading errors ; this provided three readability indices. Sentences were rendered in one of 12 different fonts consisting of four kinds of style-Old, Standard, Modern, and UD-times three weight levels-Light, Regular, and Bold. Findings for the style suggest that the enlargement of relative size represents a tradeoff with narrowed inter-letter spacing. This means that good legibility of single letter design may not result in good readability of letters in sentences. However, the weight had a notable effect especially in small sizes. Two readability indices were predicted relatively well by participants’ acuity and stroke width. The effect of stroke width had a ceiling between 10 and 15 % of the letter size.

    Download PDF (1898K)