The measurement difficulties for obtaining the values of parameters of Table I and the low weight of the decor layout in the view-though index formula has suggested a new experimental approach.

Establishing the quality of vision is not an easy work. Vision is a very complex exercise even if it seems effortless. The first step of vision considers the formation of an optical image upon the retinal mosaic receptors, then the optical image passed through several sampling process and reach the brain where perception and cognition acts start.

The vision of a scene trough glasses with geometrical obstructions is affected by the interference of geometrical pattern at both levels of the vision process. The geometrical obstructions generate well localised diffusion effects that can play a relevant role, depending on the geometrical pattern of the obstructions, in the vision process.

First of all the geometric obstructions affect the image formation in the eye, then image cognition in brain. For these reasons it is better to analyse the problem under two different aspects: the influence of the glass pattern on the contrast sensitivity (image formation) and on the scene perception.

The influences on image formation can be easily investigated considering the principles of the Fourier optics. Fourier analysis is usually applied to optical imaging systems where a very simple method to analyse their performances is to consider the spatial modulation transfer function (MTF), i. e. the frequency response of the system to spatial variation in
luminance, of the object framed. This target has usually sine wave or square wave luminance variation.

This kind of analysis was extend to the human visual system in the latter part of the XX century and actually is well established in literature [5, 6].

A sine wave or square wave luminance variation target are used to test the contrast sensitivity at different luminance contrast levels and for different spatial frequency (of luminance variation) in cycles per degree.

If a dispersive medium is interpose, i. e. a glass with geometrical obstruction, a degradation of the MTF of the eye occurs. This degradation is directly related to the modulation transfer function of the whole system, i. e. eye plus dispersive medium (glass). The dispersive medium acts like a two-dimensional filter, which attenuate amplitude and introduce phase shift in the image. Because the MTF function of the system is related to the Fourier Transform (FT) of the image, to investigate the degradation it is necessary to calculate the FT of the image

At this stage of the analysis we are only interested in the evaluation of the interferences of the dispersive medium and the FT is studied using a CCD camera. Two simple target of know luminance variation (sine wave and square wave variation) are acquired with a scientific CCD camera peltier cooled, with the glass, with geometrical pattern, interpose on the optical path. An example of CCD acquisition a test target is shown in figure 6.

A square wave variation can be easily obtained using a source framed by a rectangular diaphragm obtaining a line function. The luminance contrast of this kind of target is also easily variable. While about the sine wave variation it is know that the variation measured along the optical axis is:

1 = Im + 10 Sin(2wTA )

where Im is the mean intensity and nl is the spatial frequency of the target.

The MTF function of the glass is obtained as ratio between the MTF of the target with and without the glass. In this kind of analysis the noise of the image plays a relevant role affecting the final results. To minimise the noise the source lighting the target is a current stabilised incandescent lamp and the CCD camera used to acquired the image is peltier cooled to minimise electronic noise, with a 14 bit AD controller. These features can assure the minimum noise for the analysis.

This approach does not required complex measurement and expensive instrumentation and can give a good evaluation of the glass performance if the geometrical measurement conditions are correctly selected, considering the actual condition of vision and perception. For these reason, after the mathematical study of the influences of obstructed glass on the vision system (image formation) it is necessary to analyse the second level of vision: the image cognition.

The image cognition and perception is a very complex field of study well established in literature [7]. In the image cognition and perception analysis, the image transmitted to the brain is filtered several time and some phenomenon of image finishing, that allows the correct interpretation of the scene, arrives.

In the case of vision trough geometrical obstructions mechanisms of attentive vision are involved. Attention enables visual processing in a wide variety of ways, one that is interesting for us is that enables binding.

The visual cognition system, using the geometrical properties of binding contours and closure of the scene, is able to connect the contours using an organizing process based on the continuity perception (Gestalt school).

In the vision trough obstructed glass the attention is lasting and sustained, a top down elaboration process scene arrives.

To understand how obstructed glass affect the vision quality at this stage it is obvious that some perceptive experiments must be carried out using selected observers.

A simple test can be performed using standard visual acuity target: Landolt rings. The observer can be exposed to the target with and without the glass interposed at different adaptation levels. In this way it will be possible to check the influences on visual acuity. Otherwise to test the influences on complex scene, more complex visual task must be used. A parameter can be the time necessary to recover a specific detail in a complex scene, or perform repeated search on two similar images, seeing one trough the glass, to find the differences.

The subjective experiments should be planned considering not only the geometrical pattern of the glass obstructions but also the behaviour in which the glass panes will be installed, taking care of the effective visual tasks of users viewing through the glass. Actually some preliminary experiments to identify the main parameters are in the development phase.

Conclusions

New advance transparent systems are on the market or in a well developed research phase. Such products aim at merging the glazing functionality with more aesthetics concepts. Printed or laminated glass pane with many different decors can have several applications, but it is important the evaluation of how the quality of vision is affected. Photometric measurements stressed that the behaviour of such systems does not differ very much from conventional glazings and the amount of the transmitted energy strongly depend on the ration of obstructed area to the total one. For the three selected sample good light transmittance was measured, at off normal incidence as well. Goniophotometric measurements confirmed the visual inspection of the sample, which means mainly regular transmission of the samples, no redirecting component and very low scattering behaviour. The view through index also stressed that decors do not strongly affect the quality of vision. As tested by the pictures survey, colours and shapes are not strongly modified by the obstructions of the glass panes.

To analyse in depth the phenomenon of the quality of vision, a new approach is introduced. It is based on the analysis of image by CCD camera. The development of a complete methodology will need lot of investigations in terms of optical measurements and subjective responses of users under fixed experimental conditions. This part will be the next phase of the research on obstructed glass panes for building and daylighting applications.