=Paper=
{{Paper
|id=Vol-2353/paper17
|storemode=property
|title=Genetic Method of Image Processing for Motor Vehicle Recognition
|pdfUrl=https://ceur-ws.org/Vol-2353/paper17.pdf
|volume=Vol-2353
|authors=Ievgen Fedorchenko,Andrii Oliinyk,Alexander Stepanenko,Tetiana Zaiko,Anton Svyrydenko,Dmytro Goncharenko
|dblpUrl=https://dblp.org/rec/conf/cmis/FedorchenkoOSZS19
}}
==Genetic Method of Image Processing for Motor Vehicle Recognition==
https://ceur-ws.org/Vol-2353/paper17.pdf
Genetic method of image processing for motor vehicle
recognition
Ievgen Fedorchenko1[0000-0003-1605-8066], Andrii Oliinyk2[0000-0002-6740-6078], Alexander
Stepanenko3[0000-0002-0267-7440], Tetiana Zaiko4[0000-0003-1800-8388], Anton Svyrydenko5 [0000-
0003-0114-8632]
and Dmytro Goncharenko6[0000-0001-8973-3742]
1,2,4,5,6
Dept. of Software Tools, Zaporizhzhia National Technical University, Zaporizhzhia
69063, Ukraine
evg.fedorchenko@gmail.com, olejnikaa@gmail.com,
alex@zntu.edu.ua, nika270202@gmail.com,
antonsvyrydenko@gmail.com, gdimk99@gmail.com
Abstract. An analysis of scientific articles on similar topics was conducted; it
was de-termined that the best recognition accuracy rate is achieved by using
convo-lutional neural networks. Modifications of a simple genetic algorithm
(Alfa-Beta, Alfa-Beta Fixed, Fixed) were developed. The implementation of
the program for recognition of road users (cars, bicycles, pedestrians, motorcy-
cles, trucks, etc.) was developed. Also, a comparison was made between the use
of modifications of a simple genetic algorithm and the best approach for solving
the problem of road user recognition. The purpose of the research conducted
was to find an optimal approach for solving the problem of road user recogni-
tion, since the system that hasn’t been implemented yet can recognize road us-
ers accurately. It was found that the improved Alpha-Beta modification is the
best approach from the considered ones used to solve the problem. This modifi-
cation allowed getting the best accuracy rate in less time selection, in compari-
son with other base modifications and simple genetic algorithm. The obtained
results have a high practical value, since the developed modification allows op-
timizing the process of selection of values in other subject areas.
Keywords: Pattern Recognition, Genetic Algorithm, Evolutionary Algorithm,
Neural Networks, Python, OpenCV, Keras.
1 Introduction
The theory of pattern recognition exists as a branch of informatics and related disci-
plines. It develops methods for the classification and identification of objects of vari-
ous nature: signals, situations, objects that are characterized by a comprehensive
number of certain features [1]. The problem of object recognition can also be a pur-
pose of interdisciplinary researches - including work on creation of artificial intelli-
gence. It is also often used in solving practical problems in the field of computer vi-
sion. When setting the classical problem of object recognition, it is necessary to apply
a mathematical language based on logical thinking and mathematical principles. In
contrast to this approach, there are methods for recognizing objects using machine
learning and 6 artificial neural networks, formed not so formalized approaches to
recognition, and show not worse, but in some cases, a much better result.
One of these areas is road safety, which includes many narrow areas: number-plate
recognition, traffic congestion determination, recognition of road users in self-driving
cars, etc.
However, for today, problems of recognition quality are not completely solved. For
example, the vulnerability was detected in the Tesla Autopilot system: the system in
most cases cannot recognize a bicycle. It identifies a bicycle as a person or a small
car; accordingly, the system can make decisions that may endanger cyclist’s life [1].
Vulnerabilities in recognizing road users in systems such as Mazda Smart City
Brake Support [2], Waymo Autopilot [3] are also known.
It can be said that recognizing bicycles and other road users is a topical task. Rec-
ognition of road users can be used in the system of control and / or traffic safety or it
can be used to study the reasons of constructing specialized bicycle lanes, etc.
2 Analysis of literary data and problem statement
This work [4] presents a study of solving the problem of bicycle recognition on a
video stream by combining the following approaches: Histogram of Oriented Gradi-
ents (HOG) [5, 6], Support Vector Machine (SVM) [7], cascade classifier (Violi-
Jones method) [8, 9].
Presented results of performing bicycle recognition in different weather condi-
tions, and it is argued that the developed approach can be used in systems of real time
recognition.
However, using of a combination of HOG and SVM methods is not optimal, since
it allows you to recognize large enough objects that are not always present in video in
real time and due to the fact that it takes a lot of time to calculate complex parameters.
Using a cascaded classifier is more optimal in terms of recognition time. However,
this method has errors in recognition - because of the application of different scales
and the size of the scanning window, one object can be recognized as two.
In general, low image quality, weather conditions and various lighting conditions
make precise recognition more complicated for the methods used. The approach de-
veloped is complex and complicated, and each method requires sufficient computa-
tional resources that effect on results.
An expedient approach to the implementation of the bicycles recognition would be
the use of neural networks.
Thus, in [10], the study of the analysis of the flow of machines on the CCTV
(video surveillance systems) images, based on the solution of the problem of recogni-
tion of machines with the help of neural networks Faster R-CNN (the faster region-
based convolutional neural network) [11] and SSD (single shot multibox detector)
[12].
In general, SSDs are capable of quickly recognizing, but are often mistaken for
small objects, while Faster R-CNNs are slower than SSD, but can provide more preci-
sion.
Work aim is to test the specified trained neural networks on images of low quality
and under different weather conditions. As a result of the study, it was found that the
considered neural networks are not capable of recognizing all the machines, and in the
case of bad weather conditions (rain, snow, fog), the accuracy indicator falls signifi-
cantly. Therefore, the study of neural networks to detect vehicles should use more
diversity training set. In addition, the study implements the recognition of only cars,
although the concept of "traffic" covers several types of transport (motorcycles,
trucks, etc.).
In [13], the solution of the machine recognition problem by studying a convolu-
tional neural network [14] is considered, which is based on reinforced learning mis-
take and samples of error-prone probes. That is, in the training of the convolutional
neural network, it is proposed to use the training errors of the current stage as training
data in the next stage. A comparison of histogram oriented gradient recognition with
convolutional neural networks, one of which was taught in the standard way, and the
second - by the proposed authors, was made, which resulted in the proposed approach
showed better performance.
However, there are no examples of car recognition, because of the small amount
of study sample used in the study, one cannot say with complete certainty that a sig-
nificant improvement in the training of the firewall neural network has been achieved.
In [15] the decision of the recognition problem among the crowd of each pedes-
trian separately is considered. The main purpose of the study is to recognize pedestri-
ans in black and white images in various situations, the pedestrian's identification of
the image, even with partial overlap, and with high accuracy, localize it. Also, the
goal is to determine how many pedestrians are presented in the image.
The approach developed in this article is based on the scale-independent expan-
sion of the Implicit Shape Model (ISM).
Authors succeeded in realizing all goals, but the accuracy of pedestrian recogni-
tion is 71.3% and is not satisfactory.
Also, it's worth noting that the developed method works only with images. How-
ever, the method has a great potential and can, as a result of further development,
develop towards image recognition and tracking.
In [16] we consider the solution of the problem of pedestrian recognition by com-
bining several neural networks.
The proposed system consists of two subsystems: the main pedestrian detection
system for generation of all detections and a system of semantic segmentation to im-
prove the results. Pedestrian Detection system in turn consists of a generator "candi-
dates" for pedestrians and classification system. In order to provide more information
about the coordinates of the object of the classification network, it is proposed to use
the new method of soft tokens, which relates the object to all classes. For the imple-
mentation of the classification system the idea of "combined training" was imple-
mented. Also, the technique of "soft rejection" is proposed for combining the conclu-
sions of all networks of classification and the network of semantic segmentation with
the conclusion of the network of "candidates" generation.
Pedestrian recognition has been tested on four popular image sets: Caltech Pedes-
trian set, INRIA set, ETH set, KITTI set. The first three sets received the highest rec-
ognition accuracy. Also, it indicates the significant speed of the developed approach.
In [17] the implementation of pedestrian recognition is considered, using the com-
bination of the method of histogram oriented gradients, the method of reference vec-
tors and the convolutional neural network.
The purpose of work is to construct a semantic areas of interest in order to get the
foreground object is to reduce the errors associated with error detection background.
First, using a convolutional neural network trained by the Caltech Pedestrian set,
the thermal map is generated by the input image. Further, semantic interest regions
are extracted from the thermal map by morphological image processing. In the rest,
semantic interest regions share the whole image on the background and foreground, to
facilitate the work of the detectors who make the decision.
It is noted that with the help of semantic regions of interest, the work of detectors
to varying degrees improves.
However, the use of a combination of HOG and SVM methods is not optimal, as it
allows you to recognize large enough objects that are not always present in video in
real time and due to the fact that it takes a lot of time to calculate complex parameters.
In work [18] the implementation of the system of recognition of vehicles belong-
ing to 7 classes (motorcycle, car, pickup, bus, truck, truck with a trailer, truck with
several trailers) is considered on the CCTV video. The Deep Convolutional Neural
Network (DCNN) was used in the implementation.
The proposed system algorithmically consists of two tasks: localization and classi-
fication. Initially, the localization is performed, by generating regions independent of
the class, for each frame, and then using a deep convolutional neural network to pro-
duce feature descriptions for each region. Finally, using the Support Vector Machines
(SVM) method, for each region, the resulting description descriptions are compared
with the templates, and a conformity assessment and classification are performed.
The accuracy of vehicle recognition differs for each class, but in general, the accu-
racy is within the range of 92 - 95%. The accuracy of the recognition significantly
decreases when downloaded traffic, rain and fog, with low video quality.
The disadvantage of this development is the use of a deep convolutional neural
network, since its calculation requires a lot of computational resources, which is not
acceptable. The use of this neural network is possible with the presence of a powerful
CPU (CPU) or graphics processor (NVIDIA GPU), in conjunction with such artificial
intelligence environments as CUDA, Caffe, Torch.
Also, the system detects recognition errors on a small scale, then the object of one
class is recognized as an object of another class, for example, the "truck" is recog-
nized as a "truck with a trailer", and so on.
The analysis [4, 10, 13, 15-26] in the subject area suggests that conducting research
on the implementation of the system traffic participant’s recognition and increasing
the accuracy of recognition is a rather topical task.
3 Purpose and objectives of the research
The purpose of the study is to develop a program for solving the problem of road
users recognition - cars, bicycles, pedestrians, motorcycles, trucks. Also, it is neces-
sary to develop an evolutionary method that would allow to obtain the best indicators
of recognition accuracy when selecting the parameters of teaching neural networks at
lesser time.
To achieve this goal, the following tasks were solved:
1. Conduct an analysis of existing methods for solving the pattern recognition prob-
lem.
2. To develop methods of evolutionary search for selection of parameters for teaching
neural networks.
3. To implement the program realization of developed methods of evolutionary
search.
4. To conduct the study of the effectiveness of the developed method.
4 Materials and methods of studying the influence of
parameters in the training of neural networks
4.1 Equipment used in the experiment
The study of the influence of the values of parameters in the training of neural net-
works on the accuracy and time of training was carried out on a computer with an
Intel Core i5 7400 processor with a clock speed of 3 GHz, with a memory capacity of
8 GB.
To study neural networks, a study sample of 8,000 images was divided into 5
classes and 3 sub-sets: educational (1000 images), test (200 images), verifying (400
images).
4.2 Methods of determining the influence of parameters of teaching neural
networks
The study of the effect of values of the parameters of training neural networks was
carried out using a simple genetic algorithm and its modifications.
The sequence of actions of a simple genetic algorithm is as follows: generation of
an initial population Ps consisting of N individuals occurs:
Ps=( I 1,…,I N ) , (1)
where I - the individual.
Each individual I (its genes) is encoded by certain values from the set of all pa-
rameters of the set Q. Let the parameters of the neural networks have a set of parame-
ters Q:
Q=(n j ,m,f act ,f opt ) , (2)
where n - number of neurons of a certain layer,
j = 1, ..., m,
m - number of layers,
fact - activation function
fopt is a function of optimization.
Then the individual IN of the initial population Ps can be encoded by formula (3).
Q(ni )=R[32 ,64 ,…,nmax ],
Q(m)=R[1,2 ,…,mmax ], (3)
Q( f act )=R[relu,elu, tanh ,sigmoid ],
Q( f opt )=R[adam,sgd,ada max ,nadam,rmsprop,adagrad,adadelta],
where i = 1, ..., m,
R - function of choice of a random variable,
nmax - maximum number of neurons,
mmax - maximum number of layers.
I N =R(Q). (4)
Next, to evaluate the optimality of each individual I of the population of the PS,
the fitness function ffit is calculated:
f fit ( Ps )= max( I i ) , (5)
where i=1,...,N.
The next step is to check if the conditions for completing the search for the
optimal solution are not fulfilled. Such conditions can be the number of generations
N, the value of the fitness function ffit, runtime, and so on. In the example of selecting
parameters for teaching neural networks, the value of fitness function is the accuracy
of the trained neural network. If the conditions are fulfilled, it is necessary to
complete the execution of the algorithm and output the results, the most optimal
generation of the individual.
Then the selection of specimens is performed in a new population Pn: individuals
are ranked according to the fitting indicator, 40% of the most adapted individuals (T)
are selected in a new population. Also randomly selected 10% less fitting individuals
(L). Another 50% of the new population Pn is obtained by crossing the selected
individuals (K).
Pn=T+L+K. (6)
Intersection (C) of individuals has the following algorithm: subsets of B and W
randomly selected individuals on M and F having a specific set of genes:
M=((n11,…,nm1 ) ,m1,f act
1 1
,f opt ), (7)
F=((n12 ,…,nm2 ) ,m 2 ,f act
2 2
,f opt ).
Then the crossing can be expressed as follows:
C=( M F ). (8)
The result of the crossing of individuals is two descendants K1, K2, which can be
written as follows:
K 1=( R[(n11,…,nm1 ) ,(n12 ,…,nm2 )],R[m1,m 2 ],R[ f act
1 2
,f act 1
],R[ f opt 2
,f opt ), (9)
K 2=( R[(n11 ,…,nm1 ) ,(n12 ,…,nm2 )],R[m 1 ,m 2 ],R[ f act
1
,f act2 ],R[ f opt
1 2
,f opt ]) ,
where R - function of the choice of a random variable.
This approach of crossing individuals is called uniform cross-linking, since the
coding of the offspring randomly selects the values for each gene from each ancestor.
For the received descendants the mutation operator μ is applied according to (3):
μ( K )=R[ R (ni ) ,R (m) ,R ( f act ) ,R ( f opt )]. (10)
Then, after applying the mutation operator, the descendant can be written as
follows:
K μ=((ni ,…,nm ) ,m,f act ,f opt ). (11)
After applying the mutation operator descendants included the new generation Pn.
The bursts and mutations are held until a new generation of Pn of size N is created:
Pn=((T+L) ,K j ) , (12)
where j = 1, ..., (N- (T + L)).
Next, the ffit fitness function for each new generation Pn (5) is calculated.
It is checked whether the conditions for completing the search for an optimal
solution were not fulfilled. If the conditions are fulfilled, it is necessary to complete
the execution of the algorithm and output the results, the most optimal generation of
the individual. If the conditions are not fulfilled - continue the implementation of the
algorithm, proceed to the formation of the next generation Pn+1 based on the
generation Pn.
The proposed modifications to a simple genetic algorithm are as follows:
1) modification Alfa-Beta: for crossing in each generation, a different number of
couples for crossing is selected, in which one individual refers to the most suitable
one and the other to the least adapted individuals. Also, random mutations (base and
double) or one mutation (base) can occur: Monte Carlo method generates a random
number 0 or 1. If 0 falls, then one mutation arises if there is one mutation - there are
two mutations.
The sequence of actions of this modification of a simple genetic algorithm is
similar to its basic version, but it has some differences. At the stage of selection of
individuals to a new population of Pn individuals are ranked according to the fitness
parameter, then the number of pairs is determined randomly - a certain number of the
most adapted individuals, and the same number of the least adapted.
The most suitable individuals form a subset of B, the least adapted - subset of W.
Both subsets are included in the set of pairs V. The number of individuals that can be
chosen in pairs is in the range of 20-60% of the total number of individuals. The rest
of the new population Pn is obtained by crossing the selected individuals (K).
Pn=(V,K j ) , (13)
where j = 1, ..., (N-V).
Also, in the proposed modification, two (μ1, μ2) or one mutation (μ) may occur
randomly, whereas in the basic version of the algorithm, one mutation arises
randomly. Moreover, in a situation when there are two mutations - one gene can
mutate twice. After applying the mutation operator descendants included the new
generation Pn. Crossing and mutation are held as long as we create the next generation
of Pn size N (13).
2) modification Alfa-Beta fixed: the number of couples for crossing is selected in
each generation, with one person referring to the most suitable one and the other to
the least adapted individuals. Also, random mutations (base and double) or one
mutation (base) can occur: Monte Carlo method generates a random number 0 or 1. If
0 falls, then one mutation arises if there is one mutation - there are two mutations . A
fixed point of crossing is established - the first half of the genes participate in the
crossing - the genes responsible for the number of neurons in the layers, the values of
other genes are always passed on to the descendants of one of the individuals.
The sequence of actions of this modification is similar to the sequence of Alfa-
Beta modifications, however, the crossing operation is excellent: according to formula
7, subsets B and W are randomly selected by the individual M and F having a certain
set of genes. Intersection can be expressed as:
1 1 1
C* =(( M F ) , F ). (14)
2 2 2
The result of the crossing of individuals is two descendants K1, K2, which can be
written as follows:
K 1=( R[(n11 ,…,nm1 ) ,(n12 ,…,nm2 )],m 2 , f act
2 2
,f opt ), (15)
K 2=( R[(n11 ,…,nm1 ) ,(n12 ,…,nm2 )],m 2 , f act2 ,f opt
2
),
where R is a random variable selection function,
m2, f 2act, f 2opt - the value of the genes transmitted from the individual F.
3) modification Fixed: a fixed point of crossing is established - half of the genes
are involved in the crossing - the genes responsible for the number of neurons in the
layers, the values of other genes are always passed on to the descendants of one of the
individuals. Also, in the mutation stage randomly, there are two mutations (base and
double) or one mutation (base): the Monte Carlo method generates a random number,
0 or 1. If 0 falls, then there is one mutation, if there is 1 - occurs two mutations. The
sequence of actions of this modification is similar to the sequence of Alfa-Beta modi-
fication, but there is a difference: the selection of individuals to a new population
occurs both in the basic version of the genetic algorithm (6), and the crossing occurs
as in the second modification (14), (15).
5 Experiments and results of research on the use of
modifications of a simple genetic algorithm
There were 4 experiments in which the genetic algorithms and the simple simple ge-
netic algorithm were considered. All experiments were carried out with the same pa-
rameters: the number of generations - 5, the size of the population - 20, the mutation
rate - 0.01. Input data is a sample of 5-digit images (pedestrian, bicycle, motorcycle,
auto, truck), in which each class has 1600 images.
To modify Alpha-Beta and a simple genetic algorithm, the elite coefficient was
also set at 0.4 and the randomly selected individuals were 0.1.
In the figures below, graphs of parameter selection parameters are presented in
the training of the neural network by the proposed modifications of the genetic
algorithm. After selecting the parameters of learning by the genetic algorithm and its
modifications, the recognition accuracy for the best neural network and the selection
time were obtained. These results are shown in the Table 1. Figure 1 shows a graph of
the mean accuracy for each generation of selection parameters for training neural
networks. Figure 2 shows a chart of indicators for the 5 best individuals of the last
generation of parameter selection for each neural network.
Fig.1. Schedule of mean accuracy for selection of parameters for teaching neural networks
Fig.2. Schedule of accuracy indicators of the last generation when selecting parameters for
teaching neural networks
Table 1. Results of selection of parameters using a simple genetic algorithm
Accuracy on test Accuracy
GA version Time of training
data of auto-testing
8 days, 8 minutes,
Basic 79.45 % 85.89 %
39 seconds
modification 7 days, 7 hours, 16
82.12 % 86.90 %
Alpha-Beta minutes,43 seconds
modification 10 days, 12 hours,
80.02 % 85.89 %
Alpha-Beta fixed 4 minutes, 58 seconds
modification 5 days, 22 hours,
78.48 % 85.48 %
Fixed 27 minutes, 5 seconds
Based on the results presented in table 4, we can conclude that the use of the Al-
pha-Beta modification of the genetic algorithm is the best approach to achieving
higher recognition accuracy in less time.
6 Discussion of the results of research on the use of
modifications of a simple genetic algorithm
In the Fig. 4 and 5 are graphs of the execution time (in minutes) of the proposed
method on computer systems, which depends on the number of cores involved. It can
be seen from the graphs that the proposed method has an acceptable degree of paral-
lelism and is effectively performed on both MIMD and SIMD systems. This way, the
IPME cluster was able to reduce the method execution time from 1565 minutes (on
one core) to an acceptable 147 minutes on 16 cores. On the ZNTU the computing
system, the method execution time was reduced from 1268 minutes on a single core to
110 minutes on 16 cores. The differences in the performance of the systems are due to
their architectural features: in the cluster cores are connected by means of the Infini-
Band communicator, and in the multi-core computer they are located on a single chip,
which explains the smaller impact of overhead (transfers and synchronizations). In
addition, the processor in multi-core computer supports Turbo Boost technology [25-
32], making the time of the method execution on the single core much less than the
execution time on the core of the cluster that does not support this technology.
As a result of the study, a modification of a simple genetic algorithm - Alfa-Beta
- was developed. Using this modification allows you to speed up the selection of the
parameters of training neural networks, and increase the accuracy. It has been shown
that the increase in the number of mutations and selection in pairs of different indi-
viduals provides a greater variety of gene combinations, which leads to better per-
formance in less time.
Fig. 3-7 shows examples of recognition using a trained neural network.
Fig. 3. Example of recognizing the object "bicycle"
Fig. 4. Example of recognizing the object "car"
Fig. 5. Example of recognizing the object "motorcycle"
Fig. 6. Example of recognizing the object "pedestrian"
Fig. 7. Example of the recognition of the object "truck"
Based on the recognition results, we can conclude that a trained neural network is
able to confidently determine the relevance of an object to a particular class.
7 Conclusion
An analysis of existing methods of pattern recognition from scientific articles on
similar problems was carried out. It was determined that the best recognition accuracy
rate is achieved using convolutional neural networks.
Three modifications of a simple genetic algorithm (Alfa-Beta, Alfa-Beta fixed and
Fixed) were developed for selecting parameters of training neural networks.
A program using the developed modifications of a simple genetic algorithm for
selecting parameters of training neural networks and recognition of road users was
implemented. When applying the developed modifications, the following results were
obtained:
- Alpha Beta - accuracy 86.90%, selection time - 7 days, 7 hours, 16 minutes, 43
seconds;
- Alpha Beta fixed - accuracy 85.89%, selection time - 10 days, 12 hours, 4
minutes, 58 seconds;
- Fixed - accuracy 85.48%, selection time - 5 days, 22 hours, 27 minutes, 5
seconds.
Comparison of the rates of a simple genetic algorithm was conducted; it was
determined that the Alpha-Beta modification is the best approach to recognize road
users since this modification allowed better accuracy (86.90%) for shorter selection
time (7 days, 7 hours, 16 minutes, 43 seconds).
Acknowledgment
The work was performed as part of the project “Methods and means of decision-
making for data processing in intellectual recognition systems” (number of state regis-
tration 0117U003920) of Zaporizhzhia National Technical University.
References
1. Lugovskaya, A. M.: Artificial neural networks in medical diagnostics [Iskusstvennyie ney-
ronnyie seti v meditsinskoy diagnostike]. Computer systems and networks: proceedings of
the 54th scientific conference of postgraduates, undergraduates and students 2018, vol. 1,
pp. 182 – 183. BGUIR, Minsk (2018).
2. Goold, S.D., Lipkin, M.Jr.: The doctor-patient relationship: challenges, opportunities, and
strategies. Journal of General Internal Medicine (1), 23–33 (1999).
3. Van Tuc, N.: Approximation contexts in addressing graph data structures. University of
Wollongong Thesis Collection, 30–55 (2015).
4. Barkoulas, J. T., Baum, Ch. F., Travlos, N.,: Long Memory in the Greek StockMarket.
Applied Financial Economics, vol. 10, no. 2, 177–184 pp. (2000).
5. Dalal, N. Histograms of Oriented Gradients for Human Detection. CVPR, 886–893 (2005).
6. Schmidhuber, J., Wierstra, D., Gagliolo, M., Gomez, F.: Training Recurrent Networks by
Evolino. Neural computation. vol. 19(3), 757–779 pp. (2007). doi:
10.1162/neco.2007.19.3.757.
7. Bland, J.M., Altman, D.G.: Statistics notes: measurement error. BMJ. 312 (7047): 1654
(1996). doi:10.1136/bmj.312.7047.1654
8. Viola, P., Jones M.J. Rapid Object Detection using a Boosted Cascade of Simple Features.
Proceedings IEEE Conference on Computer Vision and Pattern Recognition, 1–9 (2001).
9. Viola, P. Robust real-time face detection. International Journal of Computer Vision, 57, 2,
137–154 рр. (2004).
10. Peppa M. V., Bell D., Komar T., Xiao W. Urban traffic flow analysis based on deep learn-
ing car detection from CCTV image series. Int. Arch. Photogramm. Remote Sens. Spatial
Inf. Sci., 499–506 рр. (2018).
11. Ren S., He K., Girshick R., Sun J.. Faster R-CNN: Towards real-time object detection with
region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelli-
gence, 39, 6, 1137–1149рр. (2017)
12. Liu W., Anguelov D., Erhan D., Szegedy C., Reed S., Fu C., Berg A. SSD: Single Shot
MultiBox Detector. Computer Vision ECCV, 21–37 (2016).
13. A CNN Vehicle Recognition Algorithm based on Reinforcement Learning Error and Er-
ror-prone Samples, http://iopscience.iop.org/article/10.1088/1755-1315/153/3/032052/pdf .
14. Yarymbash, D., Yarymbash, S., Kotsur, M., Divchuk, T.: Analysis of inrush currents of
the unloaded transformer using the circuitfield modelling methods. Eastern-European
Journal of Enterprise Technologies, vol. 3, № 5 (93), pp. 6-11 (2018). doi: 10.15587/1729-
4061.2018.134248
15. Boden M.: A guide to recurrent neural networks and backpropagation. Halmstad Univer-
sity, pp. 1–10 (2001).
16. Yarymbash, D., Yarymbash, S., Kotsur, M., Divchuk, T.: Enhancing the effectiveness of
calculation of parameters for short circuit of threephase transformers using field simulation
methods. Eastern-European Journal of Enterprise Technologies, vol. 4, № 5 (94), pp. 22-
28 (2018). doi: 10.15587/1729-4061.2018.140236.
17. Whitley D. Genetic Algorithms and Neural Networks. Genetic Algorithms in Engineering
and Computer Science, pp. 203-216, (1995).
18. Yarymbash, D., Yarymbash, S., Kotsur, M., Divchuk, T.: Analysis of inrush currents of
the unloaded transformer using the circuitfield modelling methods. Eastern-European
Journal of Enterprise Technologies, vol. 3, № 5 (93), pp. 6-11 (2018). doi: 10.15587/1729-
4061.2018.134248
19. Oliinyk, A., Subbotin, S., Lovkin, V., Leoshchenko, S., Zaiko, T.: Development of the in-
dicator set of the features informativeness estimation for recognition and diagnostic model
synthesis. 14th International Conference on Advanced Trends in Radioelectronics, Tele-
communications and Computer Engineering (TCSET 2018), pp. 903–908, (2018). doi:
10.1109/TCSET.2018.8336342.
20. Leoshchenko, S., Oliinyk, A., Subbotin, S., Zaiko, T.: Methods of semantic proximity ex-
traction between the lexical units in infocommunication systems. 4th International Scien-
tific-Practical Conference Problems of Infocommunications. Science and Technology (PIC
S&T 2017), 7–13 pp. (2017). doi: 10.1109/INFOCOMMST.2017.8246137.
21. Leoshchenko, S., Oliinyk, A., Subbotin, S., Gorobii, N., Zaiko, T.: Synthesis of artificial
neural networks using a modified genetic algorithm. Proceedings of the 1st International
Workshop on Informatics & Data-Driven Medicine (IDDM 2018), pp. 1-13 (2018). dblp
key: conf/iddm/PerovaBSKR18
22. Stepanenko, O., Oliinyk, A., Deineha, L., Zaiko, T.: Development of the method for de-
composition of superpositions of unknown pulsed signals using the secondorder adaptive
spectral analysis. EasternEuropean Journal of Enterprise Technologies, vol. 92, issue 2/9,
pp. 48–54, (2018). doi: 10.15587/1729-4061.2018.126578
23. Alsayaydeh, J.A., Shkarupylo, V., Hamid, M.S., Skrupsky, S., Oliinyk, A.: Stratified
Model of the Internet of Things Infrastructure, Journal of Engineering and Applied Sci-
ence, vol. 13, issue 20, pp. 8634-8638, (2018). doi: 10.3923/jeasci.2018.8634.8638.
24. Shkarupylo, V., Skrupsky, S., Oliinyk, A., Kolpakova T.: Development of stratified ap-
proach to software defined networks simulation. EasternEuropean Journal of Enterprise
Technologies, vol. 89, issue 5/9, pp. 67–73 (2017). doi: 10.15587/1729-4061.2017.110142
25. Kolpakova, T., Oliinyk, A., Lovkin, V.: Improved method of group decision making in ex-
pert systems based on competitive agents selection. IEEE First Ukraine Conference on
Electrical and Computer Engineering (UKRCON), Institute of Electrical and Electronics
Engineers, pp. 939–943, Kyiv (2017). doi: 10.1109/UKRCON.2017.8100388
26. Oliinyk A., Fedorchenko I., Stepanenko A., Rud M., Goncharenko D.: Evolutionary
method for solving the traveling salesman problem. Problems of Infocommunications. Sci-
ence and Technology : 5th International Scientific-Practical Conference PICST2018,
Kharkiv, 9–12 October 2018, Kharkiv, Kharkiv National University of Radioelectronics,
pp. 331-339 (2018). doi: 10.1109/INFOCOMMST.2018.8632033