We usually evaluate picture quality by S/N ratio or human eye observation. S/N ratio represents a fidelity of reconstructed picture and it doesn't represent differences of the distortion. Therfore it doesn't always good at picture quality evaluation. In this method, reconstructed pictures in which the same distortion appears have simular diagram patterns. This shows these diagrams are useful to evaluate picture quality, because it represents the error rate in its size and the difference of distortion in its pattern.
This paper describes the different size comparison method for evaluation of converted facsimile picture quality. It is clarified that picture quality degradation due to facsimile size conversion can be extracted by the different size comparison method. This method can be applied to the evaluation of conversion algorithm and communication system which has both transmission impairment and conversion distortion.
In experiment I, a triangle wave was used to produce a vertical line. The critical frequency for non-repetitive images of the line during saccade was higher than CFF obtained by fixating on it. In experiment II, the brightness of the vertical line produced by 60KH_Z triangle wave was modulated by a square wave. On the high contrast condition the results were the same with experiment I, but on the low contrast condition any repetitive images were not observed even during saccade. These results were discussed in term of saccadic suppression and visual direction constancy.
An acceptable luminance levels of a character color and a back-ground color were adjusted by twenty six subjects. A preferable color-on-color combinations were estimated by method of pair comparisons. Visual fatigue was measured after prolonged VDT work with a positive and negative presentation mode of VDT screen. The results show that the acceptable luminance value is about 30 cd/m^2,a dark-background VDT are prefered to a light-background VDT, and a lesser degree of visual fatigue is induced with the positive presentation mode.