=Paper=
{{Paper
|id=Vol-2744/paper65
|storemode=property
|title=Psychovisual Perception Scale Based on a Neural Network
|pdfUrl=https://ceur-ws.org/Vol-2744/paper65.pdf
|volume=Vol-2744
|authors=Vladimir Budak,Ekaterina Ilyina
}}
==Psychovisual Perception Scale Based on a Neural Network==
https://ceur-ws.org/Vol-2744/paper65.pdf
Psychovisual Perception Scale Based on a Neural
Network
Vladimir Budak [0000-0003-4750-0160] and Ekaterina Ilyina [0000-0003-0783-0931]
Moscow Power Engineering Institute (National Research University),
Krasnokazarmennaya 14, Moscow, Russia
budakvp@gmail.com,kitesika@gmail.com
Abstract. The purpose of this article is to construct a psychophysical scale of
visual perception from lighting scene based on a direct propagation neural net-
work using for assessment of real or synthesized images with spatial brightness
distribution.
Visual perception assessments of different scenes were obtained for 10 ob-
servers at the experimental installation of the Department of lighting engineering
of the MPEI (NRU). These results were checked and found out agreed with the
numerical scale of visual perception proposed by Lekish and Holladay. Neural
network was trained to predict a sensation at the level of 40-70%, depending on
the scale category. For more careful prediction level in each of 5 categories of
scale a new experiment should be done with new calibration and with tested in-
structions and with more observers involved.
The novelty consists in using a neural network as an expert to assess the de-
gree of comfort of the lighting scene.
Keywords: Glare Discomfort, Scale of Visual Perception, Neural Network
1 Introduction
Currently, neural networks (NN) are widely used in image and text recognition tasks,
in medicine as diagnostic systems, in quality control systems and in many other fields
where it is impossible to consider all the conditions that affect the decision. And there-
fore, only the most important ones are considered. In lighting engineering, the problem
of psychophysical assessment of a visual perception of the spatial brightness distribu-
tion into person's field of view can be assigned to this class of tasks. Processing of
visual information can be conditionally represented as: "Stimulus – sensation – percep-
tion - reaction" where the stimulus is the spatial brightness distribution and the reaction
is the sensation of comfort or discomfort caused by stimulus.
Modern lighting calculation programs allow us to get quite realistic visualizations of
the designed lighting installations. To evaluate the quality of the brightness distribution
_________
Copyright © 2020 for this paper by its authors. Use permitted under Creative Commons License
Attribution 4.0 International (CC BY 4.0).
2 V. Budak, E. Ilyina
in terms of visual comfort using real or synthesized images, a psychophysical lighting
comfort scale (PLCS) should be built to rank all these images. PLCS could be used to
set the relationship between human sensation and the numerical values of the light
source’s and background’s brightness on the image. The probabilistic model of transi-
tion from category to category of sensation should be used in PLCS to predict with
some degree of accuracy what reaction can be expected from a lighting scene.
The purpose of this work is to construct a PLCS using empirical data obtained from
an experiment where observer's reaction the depending on the brightness of a bright
light source observed on a uniform background is evaluated when performing a typical
visual task.
Lekish and Holladay were the first who used a psychophysical assessment of visual
perception of lighting [1] and tried to find the relationship between the numerical values
of source’s and background’s brightness, and a set of human responses, expressed in
the categories of lighting comfort. They suggested a scale like as is scarcely noticeable,
is most pleasant, is still pleasant, is at limit of pleasant, is very comfortable, is still
comfortable, is less comfortable, is at the boundary between comfort and discomfort, is
perceptibly uncomfortable, is thoroughly uncomfortable, is at boundary between objec-
tionable and intolerable, is irritating and is painful. They had got an empirical formula
for value K as measure of «sensation» that looks like this:
K = log(Ls) + 0.25log(Q) – 0.3log(Lad) (1)
where Ls –brightness of source, Q – solid angle of light source, Lad – brightness of
adaptation or background for source. Each numeric value of K corresponds to one of
the thirteen categories of the scale, for example, K = 1.9 indicates a feeling on the
boundary between comfort and discomfort (BCD).
In turn, Hopkinson's research [2] showed that sensations on the scale could not be
clearly defined as values of brightness, they can only be expressed by an interval. So
the transition from scale’s category to another can be set by a probability function. For
example, Hopkinson shows that the function describing BCD has form as a sigmoid.
Thus, judge a sensation can be predicted only with a certain probability.
The scale of sensations was compared with the values of the discomfort index cal-
culated using an empirical formula for a single source of light using a system of cate-
gorical judgments [3]. Since the sensitivity of the eye changes exponentially as the
brightness level increases, the standard deviation interval also increases for each sub-
sequent interval on the scale. Based on this, the hypothesis of a normal distribution of
ratings on the scale was confirmed, which allows us to find the numerical boundaries
of the intervals of subjective categories in standard deviation units and determine the
numerical boundaries of the intervals for a certain probability value, for example,
p=0.5.
Construction of a PLCS using the probabilistic model of transition from category to
category can give a more complete description how the light environment (the bright-
ness distribution in space) affects a person. It can become a measure of the typical atti-
tude of a group of observers to lighting conditions.
Lakish’s and Guth’s experiment (in 1949) investigated the influence of various fac-
tors on the visual perception on BCD for different brightness levels of glare source in
Psychovisual Perception Scale Based on a Neural Network 3
the range from 1000 to 6000 cd/m2, and as a result, these studies contributed to clarify-
ing definition, which formed the basis of modern unfired glare ratings (UGR) [4].
An experimental at the Department of lighting engineering in MPEI (NRU) was
built to study the discomfort from glare sources with various shapes and brightness in
the observer’s field of view/ It’s similar to the Lakish’s and Guth’s installation, but has
modern light sources and a wider range of brightness from 33 to 100,000 cd/m2. Mod-
ern experiment for BDC shown that the results were agreed with the experiment of
Lekish and Guth installation [5]. Initially, the psychophysical scale proposed by Lekish
and Holladay was used to find BCD. But later a scale was reduced from 13 to 5 cate-
gories: barely noticeable, comfortable, uncomfortable, unpleasant, and painful. From
our point of view these 5 categories are enough for building of new PLCS.
From the point of view of NN, the task of PLCS’s construction is most like the class
of task «many-to-many», which include image recognition. Such a tasks have several
input parameters, and the result of the NN is the classification of objects into several
categories. When a network divides input vectors into two classes, it is sufficient to
have only one network output that takes the values «1» or «0». But in our case, the scale
of sensations has 5 categories, so the number of outputs should be equal to 5. For ex-
ample if the observer's response corresponds to a category barely noticeable then the
value should be written as "1", the rest as" 0", and this logic of recording should be
applied to each observer's response.
As a result, the problem is reduced next conditions: there are several photos of the
lighting system with a spatial distribution of brightness at the input, and there are real
estimates of the observers of these scenes corresponding to the five categories of the
scale at the output. During training, the NN will extract certain features from the input
images, collect and classify them in a certain way. After training, the network should
output the probability of an image or group of images falling into each category of the
scale. The maximum probability will determine the category of the comfort scale, that
is, the most likely reaction of typical observers to the distribution of brightness in the
room.
2 Methods
If we simplify the distribution of brightness on image to the ratio of the brightness
of light source of finite size viewed on the uniform background, then the input vector
into NN have two variables: y1 = LS and y2 = Lad.
In the future, the length of the vector y can be changed by adding, for example,
correlated color temperature of the light source (y3). Moreover, later y1 and y2 can be
used both from experiment and from an image with a spatial distribution of bright-ness
in a real photograph or modeled in a lighting calculated program. The recognizing and
classification of images tasks are well solved using convolutional NN. In the framework
of this article, we will not consider this issue.
The input and output data for training of the NN were obtained from an experiment
performed on an experimental installation at the Department of Lighting Engineering
of the MPEI (NRU). The technique consisted of assessing the sensation when a glare
4 V. Budak, E. Ilyina
source appeared in the observer's field of vision while reading the inscription. The back-
ground brightness was uniform. The adaptation brightness was taken equal to the back-
ground brightness. The experiment involved 7 observers with a back-ground brightness
of 57 and 113 cd/m2 in the dark time, and 3 observers with a background brightness of
66, 77, 95 and 119 cd /m2 in the daytime in cloudy weather. Each observer evaluated
the first sensation and registered the value of source brightness. Then the bright-ness of
the source changed (gradually increased or decreased) till the observer registered the
second sensation and these actions were repeated up to fifth sensation. Then the back-
ground brightness was changed to next.The measurement was carried out both on an
ascending and descending scale. The observer was seat on the chair at distance of 1.5
meters from the inscription "Experiment", which he read. The inscription was at the
level of the line of sight. The source with diameter 30 mm was located just above the
line of sight.
Reading of inscription imitates condition for performing visual work when glare
source can appear in the person’s field of view and cause sensation of discomfort.Since
the scale of sensations refers to the scale of verbal categories, the observer can find it
difficult to spread his feelings in a wide range. That’s why special attention was paid to
the briefing. It should be clear to the observer in terms that are familiar to him. In our
experiment the briefing was as follows: barely noticeable - the source becomes notice-
able during reading the inscription, this at the moment when the observer detect the
source at the first time; “comfortable” - it is pleasant to read the inscription and the light
source does but does not interfere with reading, “uncomfortable” - the sensation of a
“buzzing fly”, the source already attracts attention, but do not , “unpleasant” - distracts
from reading the inscription , "Unbearable" means pain-fully.
Initially, the experiment was done with LEDs covered white fabric without limiting
the angular size of the source. Later it was fixed. It is very important to limit angular
size of the source during experiment because it can affect on the results.
The installation was calibrated for each the brightness of source and background
using the Konia Minolta LS-110 brightness meter. A feature of the installation is that
the brightness measurement is carried out indirectly. The observer sets not the source’s
brightness but the voltage on the LEDs (U). The calibration curve LS = f (U) shown
that in the range from 7.2 to 7.8 V, the brightness increases from 20 to 1870 cd/m2 (see
Fig.1).
Brightness of background is set by the level of general lighting in the room. When it
turned out there is a contribution from general lighting into the brightness of source
surface. Therefore, calibration was performed for each background of brightness level
separately, in order to reduce the error at low brightness levels. This limitation affects
the ability to accurately determine the interval of brightness for sensations: barely no-
ticeable and comfortable.
The measurement was carried out at one adaptation level, then it was changed. There
was a pause between the measurements so that the observer's eyes could adapt. All
measured and registered data were converted into input and output vectors for transfer-
ring them into nprtool in Matlab. This application helps to solve the problems of clas-
sifi-cation of pattern recognition [6]. After that, a NN with direct distribution was cre-
ated. It contained several hidden layers and an output layer (see Fig. 2). All neurons
Psychovisual Perception Scale Based on a Neural Network 5
may have their own activation function; the number of neurons corresponds to the num-
ber of classes n on the last layer, in our case n = 5.
Fig. 1. Calibration curve for source brightness from the voltage on the LED at Lad = 54 cd/m 2
The training of NN was performed using the standard trainscg function, which can
train any NN if its weight, clean input signal and transfer functions have derivative
functions. To solve multidimensional logical problems, the Softmax activation function
is applied, which looks like this:
e yi
S ( y )i = n
e j
y
j =1 (2)
The function converts a vector with dimension j into a vector S with the same dimen-
sion. Each coordinate S(y)i of the resulting vector is represented by a real number in
the interval [0,1] and the sum of coordinates S(y)i is equal to 1. The layer index n is
usually omitted, therefore which implies that this is the last layer.
Fig. 2. The scheme of the neural network. (Input - input vector of dimension j = 2, Hidden -
number of hidden layers n = 5, Output - output layer with 5 classes, Output - probabilities for
each class)
6 V. Budak, E. Ilyina
Our NN has five outputs, then it uses Softmax function to produce a set of probabil-
ities for each sensation caused by the brightness ratio. To assess the reliability, we use
the surface defined as:
p = f(LS,Lad) (3)
where p is the maximum probability of each event on the PLCS. In other words, this
probability surface shows how each point in space with coordinates LS and Lad refers
to a sensation. The higher the probability value, the higher the confidence.To determine
the sensations on the PLCS, we build a sensation surface in the same space, which can
be defined as:
R = f(LS,Lad) (4)
Thus, our model of scale may represent some surface - a field of predictions for a given
initial data.
3 Results
Average brightness values obtained during the experiment for each category of the
scale were used to calculate values K using formula (1). Calculated values K were
matched with Lekish-Holliday’s scale (see Fig. 3).
Fig. 3. Value K and the corresponding sensation on the Lekish-Halladay’s scale
It can be noted that the values K for barely noticeable and still nice coincide rather
good. There is a slight discrepancy for the category of BCD. Thoroughly uncomfortable
Psychovisual Perception Scale Based on a Neural Network 7
and painful coincide at Lad = 54 cd/m2 and slightly diverge at Lad = 113 cd/m2. In this
case, thoroughly uncomfortable corresponds to the instruction unpleasant in our exper-
iment. The all values K > 2.7 are considered as painful [1]. Nevertheless, such a graph
does not provide information how the feeling can be translated into the brightness of
source and background and how to determine the width of each interval. To approxi-
mate the sensation according to the normal law using response frequencies for a five-
point scale, it’s needed to find the only numerical value for one fixed brightness of the
source and background. We constructed graphs with the distribution of response fre-
quencies for each category of the scale (see Fig. 4). Several maxima can be observed
within each category. This indicates the lack of statistical data. In barely noticeable
and comfortable the maxima are shifted toward the light source’s brightness 42 and 450
cd/m2 respectively. For uncomfortable there are two maxima 3107 and 8910 cd/m2
instead of one. Unpleasant and unbearably have a more smeared character without a
clear maximum. Presumably it could be due to insufficient data or because it was dif-
ficult to correctly separate the person’s sensation of discomfort and unpleasant. As for
the painful category, everyone has their own threshold and more statistics are needed
here.
Fig. 4. Graphs of the distribution of response frequencies for each category of the scale depending
on the brightness of the light source
All the results were used to train NN and to construct the PLCS where sensation is
a function of brightness of the source and background. Having built a multilayer artifi-
cial NN it’s important to determine whether it adequately describes the data and the
dependencies between them. As a rule the number of neurons in the hidden layer is
selected experimentally by comparing the accuracy of NN. So, the number of hidden
layers in the NN is a custom parameter. The minimum values of cross entropy (CE) and
8 V. Budak, E. Ilyina
percent of errors (the proportion of incorrectly classified examples) are preferred for
the training, test, and test sets. The smaller the error, the better result[10].
In our test, we take the number of hidden layers from 2 to 100. A graph of the error
depending on the hidden layers of the NN, has a form like a cardiogram. Anyway it was
determined that the NN with 32 layers has minimum percentage of error, and NN with
11 layers has an error is closest to the minimum.
Not only the CE parameters and the percent of errors could be used to evaluate the
quality of the NS operation, but also a Confusion Matrix is used. For each class of
observations, the results of assigning observations to a class are given. The matrix al-
lows to see whether the classifier confuses classes. The matrix columns correspond to
the predicted classes, and the rows correspond to the actual classes.
Fig.5 shows the percentage of error for NN with 32 layers. Classes from 1 to 5 mean
categories on the scale from barely noticeable to unbearable. According to the Confu-
sion Matrix, grades from 1 and 5 have a minimum number of errors 9.8% and 28.6%
respectively. These values correspond to the criterion of prediction accuracy of at least
70%. Grades 2, 3 and 4, have a lower the prediction accuracy. Let's see how these pa-
rameters change for NN with 11 layers (see Fig.5). The percentage of errors in all clas-
ses except the grades 2 and 4 decreases. In this case, accuracy for grade 2 (comfortable)
and grade 4 (unpleasant) is rather low. But for grade 3 (discomfort) it is over 60%
which is quite high. This clearly indicates that the NN with 11 layers works more ade-
quately compared with NN with 11 layers. To improve the accuracy of the prediction,
additional data from the experiment is required for at least the comfortable and unpleas-
ant categories.
Fig. 5. All Confusion Matrix for 32 and 11 layers
The shape of the Receiver Operating Characteristic (ROC-curves) is also an im-
portant indicator of the quality of the neural network. For an ideal classifier, the ROC-
curve graph passes through the upper-left corner, where the percentage of true positive
cases is 100% or 1, and the percentage of false positive examples is zero. Therefore,
Psychovisual Perception Scale Based on a Neural Network 9
the closer the curve is to the upper-left corner, the higher the predictive power of the
model. On the contrary, the smaller the curve bends and the closer it is to the diagonal
line, the less efficient.
As it can be seen at Fig.6 neural network with 11 layers has the ROC-curves that
pass closer to the upper-left corner, which means that the predictive ability of this model
is higher than for other two. At the same time, it is clearly noticeable that ROC-curves
for model with 32 layers passes below the diagonal. It means that model lie.
Let's construct surfaces p and R defined by formulas (3) and (4) for LS range from
100 to 100000 cd/m2 and Lad range from 54 to 120 cd/m2 using NN with 38 and 11
hidden layers. From Fig.7 the shapes of both surfaces change depending on the number
of layers. The surface obtained based on NN with 11 layers looks less ragged.
Fig. 6. ROC-curves depending on the number of hidden layers: 32(left) and 11 (right)
Fig. 7. Probability surfaces for PLCS for cases: NN with 38 layers (left) and NN with 11 layers
(right) respectively
The higher the accuracy of the prediction (or probability), the more NN works like an
"expert" who can predict: what kind of sensation will be caused by the ratio of the
brightness of source and the background in a typical group of observers.
10 V. Budak, E. Ilyina
4 Discussions
In this article, we reviewed the results of an experiment on MPEI’s installation and
the idea of using neural networks to construct PLCS depending on the ratio of the
brightness of source and the background was tested.
Despite that response rates (frequencies) curves (Fig. 4) have no maxima for un-
pleasant and unbearable, the probability surface p shows the greatest NN’s accuracy for
these categories. This may be because the observer can easily detect them, that’s why
NN can be training more accurately and provide classification with p>0.7. As for barely
noticeable, the surface p shows the probability p>0.5 (green and yellow zone). Com-
fortable and uncomfortable have the lowest values of p, they contain zones where p<0.5
(blue color). Thus, the resulting model has a prediction accuracy of 40-70%, depending
on the category of scale.
This work has shown that this idea can be used to build a model for psychological
assessment of the spatial distribution of brightness in a lighting scene from comfort
point of view if provided a enough experimental sample for training. Even though now
PLCS has lower accuracy than required, this work allows us to formulate criteria for
the future model of the scale and requirements for a new experiment. To provide best
predicational ability of NN the brightness of background in new experiment should be
at the range from 50 to 200 cd/m2 with step of 10 cd/m2 and a greater number of ob-
servers should be involved.
Also, instructions used in this work for calibration of the experimental installation
and for briefing of observers before starting can be used in future.
This can improve the accuracy of the input data, and therefore improve the work of
the NN as the main "expert" that assess the comfort of lighting scene using synthesized
images.
References
1. Luckiesh, M., Holladay, L.: Glare and Visibility. Transactions of the IES. 20: 221(1925)
2. Hopkinson, R.G.: Evaluation of Glare. Illuminating Engineering 52(6), 305–321 (1957)
3. Kotik, G., Matveev? A., Perejma, V., Tohadze, I.: Kategorijnye ocenki kachestva i ih vzai-
mosvyaz' na psihofizicheskoj shkale, Svetotekhnika, 3, 3-5 (1975)
4. Luckiesh, M., Guth, S.: Brightness in the visual field at borderline between comfort and
discomfort. Illuminating Engineering, 44(11), 650 (1949)
5. Budak, V., ZHeltov, V., Meshkova, T., Notfullin, R.: Ocenka kachestva osveshcheniya na
osnove prostranstvenno-uglovogo raspredeleniya yarkosti. Svetotekhnika, 3, 17-22 (2017)
6. Beale, M., Hagan, M., Demuth, H.: Deep Learning Toolbox Getting Started Guide R2019b,
MathWorks, 2019, 162