Contrast Threshold Function Research Articles

Stroboscopic effect: contrast threshold function and dependence on illumination level.

The stroboscopic visibility measure (SVM) is a method used to quantify the stroboscopic effect visibility in general illumination application. SVM has been defined previously based on a limited number of frequencies and participants. To validate and extend SVM, five perception experiments are presented, measuring the visibility threshold of light waveforms modulated at several frequencies, conducted in two different labs. A power function is fitted through the aggregated results to develop a stroboscopic effect contrast threshold function for a "standard observer," which can be used to normalize SVM. An additional experiment shows the dependency on illumination level, extending the validity of SVM to other applications.

Effect of contrast threshold function correction on NVThermIP model

With the innovation of infrared imaging technology, the performance evaluation of the infrared imaging system plays an indispensable role in the general technology. Therefore, the establishment of a comprehensive, scientific and reasonable performance evaluation model is a prerequisite for accurately predicting the performance of the imaging system, and is also an effective technology to design and develop the high performance imaging system. According to previous studies, in this paper we analyze the key problems that need to be solved urgently in the current research and the factors limiting the development of performance evaluation techniques. The main researches of this paper are as follows. 1) Through improving the deficiencies of the inherent physical effect evaluation methods, the accuracy and reliability of the performance evaluation model are enhanced. 2) In order to meet the needs of performance optimization design, the performance parameters provided by product design and production requirements must be selected correctly to achieve an organic combination of performance evaluation model and imaging system.The NVThermIP model, a widely used performance evaluation model, is slightly inadequate for guiding the optimization of the parameters of an infrared system. A more scientific and reasonable performance evaluation model is proposed in this paper, in which the contrast-threshold function of the system in the NVThermIP model is corrected by noise equivalent temperature difference based on the theory of the human-eye noise. By quantitatively analyzing the typical physical effects on infrared imaging system, the modeling theory and process of NVThermIP model are introduced in detail. The simulation results give a visual representation of evaluating the performance of an infrared imaging system. The limitations of the NVThermIP model used to guide the design and production of the system and the deficiencies of the early theoretical basis for system optimization design are analyzed. The noise equivalent temperature difference is introduced to revise and perfect the NVThermIP model combined with the theory of human eye noise. The accuracy of the newly proposed model is verified by two experiments. Experimental results show that the corrected model is more accurate in system prediction and can be used to guide the design of a new system.

Human vision noise model validation for the U.S. Army sensor performance metric

Image noise originating from a sensor system is often the limiting factor in target acquisition perfor- mance, especially when limited by atmospheric transmission or low-light conditions. To accurately predict target acquisition range performance for a wide variety of imaging systems, image degradation introduced by the sen- sor must be properly combined with the limitations of the human visual system (HVS). This crucial step of incor- porating the HVS has been improved and updated within NVESD's latest imaging system performance model. The new noise model discussed here shows how an imaging system's noise and blur are combined with the contrast threshold function (CTF) to form the system CTF. Model calibration constants were found by presenting low-contrast sine gratings with additive noise in a two alternative forced choice experiment. One of the principal improvements comes from adding an eye photon noise term allowing the noise CTF to be accurate over a wide range of luminance. The latest HVS noise model is then applied to the targeting task performance metric respon- sible for predicting system performance from the system CTF. To validate this model, human target acquisition performance was measured from a series of infrared and visible-band noise-limited imaging systems. © 2014

Effect of Monochromatic Induced Aberrations on Visual Performance Measured by Adaptive Optics Technology

The two objectives were 1) to measure visual acuity (VA) and contrast sensitivity (CS) losses induced by various amounts of individual Zernike aberrations, and 2) to examine the accuracy of image quality metrics in predicting these visual performance losses. Monocular 10 cycles/degree (cpd) and 25 cpd CS and high- and low-contrast VA were measured when introducing individual Zernike aberrations in four patients dynamically corrected using a CRX1 adaptive optics system (Imagine Eyes) and a 5.5-mm pupil. Seven levels (0, +/- 0.1, +/- 0.3, and +/- 0.9 microm) of astigmatism, spherical aberration, coma, and trefoil and four levels of defocus were induced. Several image quality metrics based on the radially averaged modulation transfer function (rMTF) and optical transfer function (rOTF) calculations were computed to attempt to predict the losses in VA and CS. Modes near the center and at the top of the Zernike pyramid decreased VA significantly more than modes near the edge or at the bottom of the pyramid. The measured CS losses were reasonably correlated to the rMTF calculated at 10 cpd (R2 = 0.87) and 25 cpd (R2 = 0.75). The high-contrast VA degradations were also reasonably predicted (R2 = 0.85) by the intersection between the rMTF and the neural contrast threshold function. The low-contrast VA losses were also well predicted (R2 = 0.88) by three rMTF-based metrics. Image quality metrics based on wavefront aberration measurements were able to predict the impact of individual Zernike aberrations on CS and VA.

Calculations and Measurements of the Visual Benefit of Correcting the Higher-order Aberrations Using Adaptive Optics Technology

Purpose Our goal was to examine the accuracy of metrics, calculated using a numerical eye model including the measurement of the monochromatic aberrations of the eye, to predict the contrast sensitivity (CS) and visual acuity (VA) visual benefits (VB) of correcting higher-order aberrations (HO). Methods We measured, on the right eyes of 25 subjects (10 myopes and 15 emmetropes) aged 21 to 56 years, the 16 c/deg CS and high-contrast VA in two conditions of aberration corrections: (i) when correcting only the defocus and astigmatism terms and (ii) when dynamically correcting all the monochromatic aberration terms up to the 5 th order. The measured VB was defined as the ratio of the performances between these two conditions of aberration corrections. Results We measured a VB of 1.25 and 1.64 respectively in term of VA and CS. We did not find any influence of age on the VB and no statistical significant difference between the myopic and emmetropic group. The contrast sensitivity VB was well correlated (r 2=0.79) with the ratio of the modulation transfer functions calculated at 16 c/deg in both conditions of aberrations corrections (i.e. MTF 16c/deg HO/MTF 16c/deg SC). The levels of correlation between various metrics and measured visual acuity VB were lower (r 2=0.30 in the better case), however the averaged VB was correctly predicted by the ratio of the intersections between the MTF and a typically neural contrast threshold function. Conclusions Metrics based on wave aberration measurements are able to predict the impact of monochromatic aberrations on CS.

高齢者の文字の読みやすさと所要照度に関するVisibility Levelの観点からの検討

The recommended illuminance in the Japanese lighting standard (JIS Z 9110) was determined based on Indow and Kawai's experimental study concerning the relationship between illuminance and readability of printed text. As the observers in the experiment were young, the recommended illuminance derived from the experiment cannot be applied directly to older people. To estimate the recommended illuminance for older people, a set of numerical analyses wasconducted. We found that the readability scale in the above experimental study correlates with the Visibility Level (VL) defined by CIE publication No.19/21. The luminance difference threshold (ΔLmin) derived from the contrast threshold function (Cth) of the CIE publication was divided into two parts (ΔLaf for foveal adaptation and ΔLeq for veiling luminance), assuming ΔLmin=ΔLaf+ΔLeq. After considering age factors in each part, the luminance difference threshold function ΔLmin*was reconstructed. Applying the ΔLmin* function to the VL analysis, the recommended illuminance for older people was estimated. The result shows that at least about 700 lx of illuminance may be recommended for older people reading 3-mm high printed characters. The visibility level analysis is a useful method for assessing readability considering age-related effects.

Effects of spatial luminance nonuniformities on visual-task performance and subjective uniformity

The objectives of the experiment described in the paper were the determination of the effects of nonuniformities on operator performance and perception, and the validation of current recommended standards. Subjects performed two tasks in the presence of nonuniformities: an objective visual search task, and a subjective magnitude-estimation task for the determination of perceived uniformity. The results indicated that the nonuniformities did not appreciably affect search performance, and that current recommendations are appropriate, although the magnitude-estimation task results indicated that the subjects were sensitive to the nonuniformities. The subjective impressions of perceived uniformity results follow the contrast threshold function of the visual system.

A model of visual threshold detection

A model of visual incremental threshold detection is presented, having the following features. 1. (1) From Crozier's Law we deduce that all thresholds, for conventional as well as for noisy images, are ultimately determined by noise. Other evidence, which is described, supports this conclusion. 2. (2) It is proposed that signal and noise are sampled by apertures at a hypothetical “decision site” in the visual cortex. These apertures, which may consist of cortical receptive fields, “match” the signal as projected on the cortex. 3. (3) For circular signals < 10 mrad, the diameter of the relevant sampling aperture is equal to that of the signal as modified by the point spread-function of the optical and neural components. The signal intensity and the noise power are therefore integrated over the modified signal diameter. The model accurately predicts the contrast threshold function over a wide range of conditions for signals <10 mrad. 4. (4) For circular signals > 10 mrad, one of three qualitative mechanisms is tentatively proposed: first, a limit to the size of receptive fields, second, a limit to the active area of receptive fields and third, the effect of lateral inhibition on the large-area signal-to-noise ratio.

Limiting factors in threshold and suprathreshold vision.

THE threshold contrast of object detail decreases with increasing size; the full curve of Fig. 1 shows the threshold contrast of circular object details as a function of their angular diameter1. To explain the form of this function some2–4 have proposed that noise is less effective in “obscuring” the larger details; others5,6 have suggested that the contrast loss in the optical and neural processes is less for the larger details. This fundamental uncertainty has not yet been resolved7. According to the first view, noise is sampled over the equivalent area of the object detail. Because a larger number (N) of random events would be integrated in a larger detail area, the relative standard deviation (N−½) would be less, and would in fact be inversely proportional to the detail diameter (Fig. 1, dotted line). According to the second view, the “obscuring” effect of the noise is constant for different diameters of object detail, and the form of the contrast threshold function results solely from variation in the contrast loss. The loss arises from imperfections in the optical and the neural processes. The contrast loss may be quantitatively described in terms of a contrast transfer function. For the present purpose, this is defined as the ratio of the subjective brightness contrast to the objective luminance contrast of the object detail, expressed as a function of detail diameter. Both contrasts refer to the centre of the detail in relation to its background.

A Consideration of the Screen Brightness Problem

The great interest that the problem of optimal screen brightness holds in motion picture engineering is proved by the numerous researches on the subject in recent years. Besides the very interesting American papers published in this Journal, some recent German works are worthy of consideration. — In 1936 Zimmermann determined the dependence of the visual effect upon the screen brightness and found that a maximum value is attained at about 14 foot-lamberts. He investigated also the influence of light distribution, and temporary brightness changes. Finally he pointed out that the time the eye needs to see all recognizable contrasts varies, according to the brightness level, between 1/3 and 1/10 second. — In 1936 Rieck verified, under conditions similar to those in cinema theaters, the character of the contrast-sensibility relation that Brodhun and Konig had found in their classical research. — Very recently H. Frieser and W. Munch reported results obtained by projecting a detail test-object. They determined the contrast threshold function under conditions very similar to those of actual projection. They did not find a material increase in the number of distinguishable contrast steps for picture brightnesses exceeding 10 foot-lamberts. — It is to be hoped that the consideration of the results of (111 these investigations will form a basis for early temporary screen brightness standardization.