Abstract—With the continuous development of network computer technology, people's visual perception ability is
enhanced, and the requirements for computer design are gradually increasing. Figure and picture design of computer
is no longer limited to the simple design of graphics or images, but is more inclined to visually convey the effect and
enhance the expressiveness and beauty of graphics and images. Based on this situation, based on data mining algorithm,
this paper puts forward the optimization strategy of figure and picture design of computer and the design of visual
sense transmitting, and discusses the specific application of computer-related design in people's practical life. In the
application of figure and picture design of computer and the design of visual sense transmitting, data mining algorithm
can not only efficiently and accurately mine the final frequent set association rules, but also meet the requirements of
efficient data mining in multi-core and heterogeneous platforms, which verifies the effectiveness and feasibility of data
mining algorithm in computer graphic images.
Keywords— Data mining; computer; graphic image design; visual communication design
Received: April 28, 2021. Revised: March 18, 2022. Accepted: April 20, 2022. Published: May 18, 2022.
1. Introduction
n the digital age, with the rapid development of network and
IT, all walks of life have accumulated a large amount of data.
Using these data to dig out potential rules to better serve life
and work has become a new trend [1]. Therefore, data mining
technology plays an important role today. Nowadays, the
general public's application requirements and demands for
computer sci & tech are increasing, and people's life and work
cannot leave the figure and picture design of computer [2].
Visual taste and visual comprehension have also been improved
with the development of computer Internet technology. More
and more products need to use figure and picture design of
computer to express the connotation and function of the product
[3]. Consumers' visual needs are also gradually increasing, and
the visual experience that the product image depicted by the
computer brings to people and the visual effect that they want
to convey to people are gradually being valued by people [4].
High-level and high-quality figure and picture design of
computer and the design of visual sense transmitting can not
only convey a more real and beautiful visual experience to
people, but also promote the progress and improvement of
people's active consumer psychology [5]. Therefore, the
analysis of figure and picture design of computer and the design
of visual sense transmitting has extremely important practical
significance for promoting the development of contemporary
computer sci & tech in my country [6].
Figure and picture design of computer is a modern
professional technology integrating imagination, creativity,
artistic sense and media [7]. Moreover, figure and picture
design of computer is a new and high-tech IT, and its rapid
development indicates the progress of the times [8]. At present,
more and more young people like innovation and have wild
ideas. Their demand and requirements for all creative products
are gradually increasing [9]. Moreover, the quality of graphics
and image design will also affect the initial influence and
purchase desire of the public on products to a certain extent.
Therefore, it is precisely because figure and picture design of
computer is more and more widely used in life that its product
level and quality will be improved [10]. Computer graphics and
image related technology can present the artistic expression
desired by designers through relevant data analysis and
information processing means [11]. More importantly, the
computer can meet the design needs of designers to a great
extent, and quantify a variety of design details to very specific
values. And allow them to compare and modify, so the graphics
and images drawn based on the computer can better reflect the
design intention and design concept that the author hopes to
convey [12]. This paper first introduces the requirements of
figure and picture design of computer and the design of visual
sense transmitting, then puts forward the optimization strategy
of figure and picture design of computer and the design of
visual sense transmitting based on data mining algorithm, and
discusses the specific application of computer related design in
people's practical life, in order to provide a reliable theoretical
reference for the development of computer sci & tech.
Computer graphic image design mainly aims at processing
images and designing pictures, involving TV movies, animation,
advertising, graphic design, architectural environment art,
A Computer Graphic Image Technology with Visual
Communication Based on Data Mining
XIAOYU YAN
Academy of Fine Arts, Shaanxi Xueqian Normal University, Xi'an710100, Shaanxi, CHINA
I
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2022.18.12
Xiaoyu Yan
E-ISSN: 2224-3488
89
Volume 18, 2022
fashion design, mechanical design, packaging and other
industries, and has been well applied in the fields of writing,
mathematics and medicine [13]. The application of computer
technology in the field of multimedia design has greatly
changed people's thinking and creative ideas [14]. The
application of computer-related technologies and software to
build models of graphic images and add and display virtual
effects greatly promoted the communication between science
and art, and brought new charm and visual effects to traditional
graphic image design methods [15]. Compared with the
traditional graphic design, the graphic images designed by
computer contain more and richer contents and elements, and
convey more comprehensive and prominent meanings and
concepts [16]. Computer graphic design not only represents the
purpose of graphic design, but also contains the future
development trend of visual communication [17]. In the
application of figure and picture design of computer and the
design of visual sense transmitting, the data mining algorithm
runs well, mining the final frequent set association rules
efficiently and accurately, meeting the requirements of efficient
data mining on multi-core and heterogeneous platforms, and
verifying the effectiveness and feasibility of the data mining
algorithm in computer graphics and images.
2. Computer graphics design and visual
communication design
2.1 The similarities and differences between
figure and picture design of computer and the
design of visual sense transmitting
Figure and picture design of computer and the design of
visual sense transmitting have similarities [18]. (1) In the
learning process of development status and development
history, they all have the basis of design and development
history. (2) Part of the knowledge involved in figure and picture
design of computer and the design of visual sense transmitting
is the same. Designers must have professional knowledge and
skills in color, graphic design and painting. (3) The computer
software used in figure and picture design of computer and the
design of visual sense transmitting is the same, including Auto
CAD, 3DStudio MAX, etc. [19].
There are also differences between computer image design
and visual communication design. (1) As for the design
background, the main reason is that they are produced at
different times. Visual communication design appeared after
World War II [20]. For figure and picture design of computer,
it comes into being after the emergence of computers. (2)
Computer graphics image design only focuses on images and
graphics processing effects, but pays more attention to the
overall effect of the works. The visual communication design
must first consider the response of the first vision to the overall
effect of the people who contact the works [21]. (3) For visual
communication design, its objective task is to increase aesthetic
feeling and dynamic, and design images with artistic features
and dynamic media images. For computer image design, its
goal task is to effectively combine static images with dynamic
images, and to design two-dimensional space and three-
dimensional space by computer [22].
2.2 The influence of computer graphics
technology on visual communication design
Visual communication design works will show different
styles due to the differences in the color, text and other factors
contained in it [23]. Visual communication design can only
stand the test of the audience if it emphasizes the integration of
emotion. Emotion is the quintessence of visual communication
design. A software platform based on figure and picture design
of computer, focusing on the parameters and rules that affect
the design [24]. Use existing technology to establish a multi-
mode collaborative working environment. A collaborative
work support platform with integrated and integrated
multimedia mode. Figure 1 shows the collaborative interaction
structure of computer graphics design.
Figure 1 Graphic and image design collaborative
interaction structure
Visual communication is mainly to design some information,
and designers can use information elements to express their
design ideas. Figure and picture design of computer is widely
used in visual communication design, which is another way for
designers to express their ideas [25]. In visual communication
design, the application of computer graphics and images can
enhance the artistic features of visual communication, and
designers can enhance the artistic effect of visual
communication works through related factors such as plane and
color. Figure 2 shows the hierarchical structure of the computer
graphic image design system for a three-dimensional virtual
scene.
Figure 2 System hierarchy of 3D virtual scene
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2022.18.12
Xiaoyu Yan
E-ISSN: 2224-3488
90
Volume 18, 2022
Advertising is different from other visual communication
factors, it has a strong sense of movement, and advertising
design can not be separated from graphic image design software,
through which pictures can be synthesized and processed to
enhance the effect of advertising design [26]. When designing
advertisements, graphics and images should be applied
integrally, especially in advertising photography. Designers can
effectively combine hand-drawn methods with graphics and
image software to integrate product features, thus enhancing
people's attraction [27]. With the constant promotion of
dynamic interaction, new media design and other concepts, the
visual communication design handled by computer has sound,
light, animation and other contents at the same time. And covers
the paper media, packaging, network image, advertising,
posters and other areas. The cooperative design flow of figure
and picture design of computer is shown in Figure 3.
Figure 3 Cooperative design operation process
In daily life, people tend to be more willing to buy products
with better visual effects. It can be said that the appearance,
color and packaging of a product will greatly affect people's
consumption desire. Generally speaking, people's first
impression of the product formation is extremely important.
The graphic image design of the product and the visual
expression effect displayed by it will convey the profound
connotation and brand culture image of the product to people in
the first time. Consumers will quickly perceive whether the
product meets their own needs through their first feeling of the
product, and quickly make a judgment on whether to buy or not.
3. Graphic image design and visual
communication design based on data
mining
3.1 Image data mining technology
Data mining technology is to automatically find out the
special relationship hidden in the data from a large number of
data. The application of data mining technology has arisen due
to the maturity of data collection technology, high-speed
processor architecture and mining algorithm. It repeatedly uses
a variety of data mining algorithms to determine patterns or
reasonable models from observation data. These data can be
stored in databases, data warehouses or other information stores.
Data mining is to find out the distance or relationship between
items, and its mining technology includes frequent itemset
search, cluster analysis, prediction query, decision tree,
regression prediction and so on.
The effective use of visual communication can expand the
field of information expression. The rise of figure and picture
design of computer has injected new vitality into visual
communication design. It is an effective way of visual
communication design. The effective combination of figure and
picture design of computer and the design of visual sense
transmitting can enhance the aesthetic feeling of visual
communication works, and then bring people a beautiful
experience. Image resources in visual communication include
many aspects, such as illustrations, drawings and illustrations.
If the picture is not processed, it will become monotonous and
boring. Therefore, in order to enhance the visual
communication effect of illustration design, the pictures can be
adjusted through figure and picture design of computer. The
data mining system structure of figure and picture design of
computer is shown in Figure 4.
Figure 4 Data mining system structure designed by computer
graphics
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2022.18.12
Xiaoyu Yan
E-ISSN: 2224-3488
91
Volume 18, 2022
The design of computer technology in the field of graphics
and images belongs to quantitative design. It can decompose the
color information and structure information in graphic images
into several independent elements that can be measured and
evaluated by numerical values. Therefore, the purpose of
graphic image design based on computer is more clear, and the
collocation between the contents is more reasonable and
prominent, and the designer can be liberated from the
traditional manual drawing process. And turn your creative
inspiration into a concrete and realizable creative process, so
that computer-aided designers can achieve and complete the
precision and design effect that traditional creative methods
can't achieve. Function-driven image data mining is to design
the driving framework of mining system according to the
specific requirements of specific applications. The data mining
function-driven model is shown in Figure 5. It consists of four
functional modules.
Figure 5 Data mining function-driven model
Since the graphic grammar relationship matrix is a triangular
matrix, each row represents the grammatical relationship
between a graphic element at the front and the graphic elements
at the back of other positions. Therefore, first calculate the
similarity coefficient of the grammatical relationship of each
row, then calculate the similarity coefficient of each column,
and finally calculate the similarity of the two graphics. The
calculation method is as follows.
(1)
An effective method for texture feature extraction is based on
gray-level co-occurrence matrix. Because, the joint probability
distribution of two gray-scale pixels that are separated by
yx ,
appearing at the same time in the image can be
represented by a gray-scale co-occurrence matrix. If the gray
level of the image is set to N level, then the co-occurrence
matrix is N×N matrix can be expressed as
khM yx ,
,
,
where the element value at (h,k) represents two gray levels of h
and the other gray level of k The number of occurrences of pixel
pairs separated by
yx ,
. In this way, the various statistics
of the gray-level co-occurrence matrix can be used as the
texture feature in content-based image retrieval for feature
measurement. The statistics of the gray-level co-occurrence
matrix used for content-based image retrieval are:
(1) Contrast (moment of inertia of the main diagonal) G:
hk
h k
mkhG 2
(2)
For coarse texture, since the value of
hk
m
is more
concentrated near the main diagonal, the value of (h-k) is
smaller at this time. Therefore, the corresponding G value is
also smaller; on the contrary, for fine texture, the corresponding
G value is larger.
(2) Energy J:
2
h k hk
mJ
(3)
This is a measure of the uniformity of image grayscale
distribution. When the numerical distribution of
hk
m
is more
concentrated near the main diagonal, the corresponding J value
is larger; on the contrary, the J value is smaller.
Autocorrelation A:
h k hk
mhkA
(4)
Autocorrelation is used to describe the similarity of gray
levels between row or column elements in an image matrix.
Entropy S:
h k hkhk mmS log
(5)
When the
hk
m
values in the gray-level co-occurrence matrix
are not much different and relatively scattered, the S value is
larger; on the contrary, when the
hk
m
values are more
concentrated, the S value is smaller.
Image retrieval based on spatial relationship features has
always been an important research direction of image data
mining, and there are many methods at present. However, it is
difficult to transform the spatial relationship of images into
quantitative measurement of image similarity. In content-based
image retrieval, it is necessary to calculate the similarity
between images according to the low-level visual features of
image content, and then retrieve images according to the
similarity. The retrieval process can also be regarded as K-
nearest neighbor search with given distance measure in feature
space. Distance can be directly used for retrieval, or it can be
transformed into similarity between 0 and 1 by a monotone
decreasing function.
Assuming that the image is represented as a d-dimensional
feature vector, given that the features of two images are
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2022.18.12
Xiaoyu Yan
E-ISSN: 2224-3488
92
Volume 18, 2022
T
d
T
dyyyyxxxx ,,,,,, 2121 、
respectively,
the cosine of the angle between them can be used as the
similarity measure:
yx
yx
yxSint
,
(6)
Cosine metric is often used in text retrieval, and it is also used
in image retrieval. The distance between two histograms can be
measured by histogram minus.
d
i
d
iii
d
iii
hyx
yx
yxD
,min
,min
,
(7)
Formula (8) is mainly used to measure the similarity between
histograms. In addition, the more frequently used Minnesota
distance is defined as:
p
d
i
p
iip yxyxD /1
1
,
(8)
When p=1, it is the block distance, also known as the
1
L
distance. When p=2, it is the Euclidean distance, also known as
the
2
L
distance. In order to distinguish the role of different
feature components in the similarity measurement, their
weighted form is often used. For example, the weighted
1
L
distance is:
d
iiii yxwwyxD
1
1,,
(9)
In addition, the secondary distance is also often used, the
main of which is the Mahalanobis distance, which is defined as:
jj
d
i
d
jiiij yxyxmMyxD
,,
2
(10)
Where M is a real symmetric matrix. If M is restricted to a
diagonal matrix, the weighted Euclidean distance can be
obtained.
At present, the commonly used technologies mainly include
similarity search, image association rule mining, image
classification, image clustering and neural network. A complete
data mining process, first of all, should pre-process the data, not
only to ensure that the amount of data is enough to make the
mining results more meaningful, but also to evaluate whether
the analysis data can be finished in unit time. In addition, it is
also necessary to purify and filter the data to filter out some
unnecessary or unknown data, so as not to cause mining errors.
3.2 Design of Computer Graphics Vision System
For people in the era of modern visual
communication, digital art has penetrated into people's daily
life. It belongs to a more popular way of artistic creation. Most
young people like to try. Using this way can release their
imagination and effectively express their feelings. Figure and
picture design of computer and visual communication design
release not only the designer's hands, but also make their brain
thinking flexible, liberate their bound imagination, no longer
limited to the skill framework, and can create unconscious and
supernatural special works.
The original image can not be directly used for image mining.
We must preprocess the original image, extract the relevant
features of the image, generate the image feature database, and
mine the image features by relying on relevant technologies.
The performance parameters of layout optimization before and
after algorithm optimization are shown in Table 1. The
comparison of topology reliability optimization simulation is
shown in Figure 6.
Table 1 Layout optimization performance parameters
before and after optimization
Before
optimization
Optimized
Number of
lines
32
41
Number of
columns
10
20
Number of
monitoring
points
320
820
Figure 6 Simulation comparison of graph layout
optimization
A key problem of multimedia data mining is the
representation of image data itself. This is also the key of image
processing and pattern recognition. Generally speaking, the
basic features of an image can be represented by colors, textures,
shapes and motion vectors. Advanced concept can be regarded
as a characteristic pattern. Advanced concept is our concern, it
may be the existence of some kind of object or the occurrence
of some kind of phenomenon. There must be a mapping
relationship between the basic features of the bottom layer and
the high-level concepts, which can be found by data mining.
The simulation comparison of reliability optimization of image
segmentation topology is shown in Figure 7.
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2022.18.12
Xiaoyu Yan
E-ISSN: 2224-3488
93
Volume 18, 2022
Figure 7 Image segmentation filtering optimization
simulation comparison
The Internet is the largest place for data production,
circulation and consumption, and there are a lot of image data
on the Internet. The data source of the system mainly consists
of two parts, one is the local image database and the other is the
Internet. Taking the statistical results of big data, the index
parameters of page evaluation, as the research object, data
clustering and information fusion are carried out to realize the
evaluation of design ability. The analysis shows that the
accuracy of design capability evaluation with this method is
high, and the utilization rate of design resources is good.
Comparison of the two analysis methods is shown in Figure 8.
Figure 8 Accuracy comparison
The results of mining also need to be further transformed into
a more understandable form, and visualization is one of the
most important processing methods. Make users participate in
interactive color matching, through evolutionary operation. The
output schemes have good satisfaction and consensus, and the
evolution process is terminated. From the evolutionary process,
the maximum fitness value may decrease, but the average
fitness shows an increasing trend, with good global
convergence and faster overall convergence speed. The
variation curve of fitness value is shown in Figure 9.
Figure 9 Changes in fitness value
Multimedia data cube presents users with a view that is easier
to understand and abstract. It can evaluate the mining results
from different levels and angles, and optimize each module
according to the evaluation results. Figure and picture design of
computer is based on user experience and pays attention to the
layout of interface. From the perspective of cognitive
psychology, following the user-centered design principle is
essentially the embodiment of the people-oriented design
concept, which is a way to convey information more effectively
and enhance users' experience and feelings. In the process of
mining, real-time human-computer interaction with experts in
various fields, users and other task-related personnel is also
needed. Appropriate manual intervention can greatly reduce the
difficulty of mining and improve the efficiency of mining.
4. Conclusion
Today, with the rapid development of sci & tech, with the
help of computer image processing software, the realization
efficiency of graphic image design and visual communication
design can be effectively improved, rich design contents can be
provided for designers, and designers can use more processing
skills and means to adjust or improve the designed works.
Moreover, the combination of graphic images and visual
communication design can also create more shocking and
infectious works of art, and constantly push people to use
modern tools to show their thinking and innovation ability to
others. The design of computer graphics and images not only
helps to improve the design of graphics and images, but also
promotes the development of computers. The research of visual
communication design has brought creative and personalized
visual appreciation works to people, and influenced people's life
and consumption concept. It can be seen that computer image
design and visual communication design are interrelated and
influence each other. Therefore, in the era of information
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2022.18.12
Xiaoyu Yan
E-ISSN: 2224-3488
94
Volume 18, 2022
development, we should improve the technical deficiencies.
Strive for the use of figure and picture design of computer
means to promote artistic development. And continue to study
the innovation of visual communication design, combine them
closely, and achieve the perfect combination of technology and
art on the basis of exerting their common advantages. The
integration of the two has high application value in people's
production and life, and the future development prospect is very
broad.
Acknowledegments
This work is supported by Research Project on Major Theor
etical and Practical Problemsin Social Science Circles of Shaa
nxi Province "Research on development model of ecological c
ultural and creative products in Qinling"(20ST-52).
References
[1] Parmaxi A, Zaphiris P. Computer-mediated communication in computer-
assisted language learning: implications for culture-centered design.
Universal Access in the Information Society, 2016, 15(1):169-177.
[2] Liu Sha. Research on Computer Graphic Image Design and Visual
Communication Design. Art Technology, 2019, 032(018):102-103.
[3] Huang Biyun. Application research of computer graphics and image
processing technology in visual communication system. Digital
Technology and Application, 2019, 037(006):107-108.
[4] Woo K L, Guillaume R, Darren B. Computer-animated stimuli to measure
motion sensitivity: constraints on signal design in the Jacky dragon.
Current Zoology, 2017(1):75-84.
[5] Hanca J, Braeckman G, Munteanu A, et al. Lightweight real-time error-
resilient encoding of visual sensor data. Journal of Real-Time Image
Processing, 2016, 12(4):1-15.
[6] Yang Hongyu, Yang Bo, Wu Xiping, et al. Research and Prospect of
Intelligent Air Traffic Control Technology. Engineering Science and
Technology, 2018, v.50(04):16-25.
[7] Wang Y, Aghaei F, Zarafshani A, et al. Computer-aided classification of
mammographic masses using visually sensitive image features. Journal
of X-Ray Science and Technology, 2017, 25(1):171-186 .
[8] Yang Jiwu. Research on the development and application of computer
data mining technology[. Information and Computers: Theoretical Edition,
2018, 418(24):34-36.
[9] Liu Falun, Xu Yu. Application of data mining in computer teaching
evaluation. Science and Technology Information, 2009(4): 1-2.
[10] Kemp, Pavlina S, VanderVeen, Deborah K. Computer-Assisted Digital
Image Analysis of Plus Disease in Retinopathy of Prematurity. Seminars
in Ophthalmology, 2016, 31(1-2):159-162.
[11] Feng Yi, Li Xujie. Research on the Application of Data Mining in
Computer Dynamic Forensics Technology. China Science and
Technology Investment, 2018, 000(015):293-294.
[12] Jiang Yu, Wang Weitao, Zhang Jiangao, et al. Data quality problems and
mining in data mining. Computer Science, 2002(12):143-144.
[13] Shan P, Sun W. Auxiliary use and detail optimization of computer VR
technology in landscape design. Arabian Journal of Geosciences, 2021,
14(9):1-14.
[14] Xiao Guangyu. The Application of Computer Graphics Technology in
Data Calculation%. Information and Computers, 2017, 000(013):124-125.
[15] Lopes DS, ParreiRa PF, Mendes AR, et al. Explicit design of transfer
functions for volume-rendered images by combining histograms,
thumbnails, and sketch-based interaction. The Visual Computer, 2018,
34(12) :1713-1723.
[16] Wang Xinghong. Design and implementation of computer management
system based on data mining. 2021(2019-2):48-51.
[17] Torres M, Qiu G. Habitat image annotation with low-level features,
medium-level knowledge and location information. Multimedia Systems,
2016, 22(6):1-16.
[18] Kim C, CC Chang, Yang C N, et al. Special Issue: Real-Time Data Hiding
and Visual Cryptography. Journal of Real-Time Image Processing, 2018,
14(1):1-4.
[19] Yang Xin. The relationship between computer graphic image design and
visual communication design. Modern Information Technology, 2020,
v.4(18):95-97+102.
[20] Burnside ES, Drukker K, Li H, et al. Using computer-extracted image
phenotypes from tumors on breast magnetic resonance imaging to predict
breast cancer pathologic stage. Cancer, 2016, 122(5):748-757 .
[21] Wending Cao. The aesthetic communication of graphic design in office
culture. 2021(2018-5):120-125.
[22] Debenedictis E P. Computer Design Starts Over. Computer, 2017,
50(8):14-17.
[23] Li Zhifeng, Duan Man. Discussion on the development and application of
computer data mining technology. Information Technology and
Informatization, 2019, 000(005):231-232.
[24] Capobianco G, Cerrone C, Placido A D, et al. Image convolution: a linear
programming approach for filters design. Soft Computing, 2021, 25(14):
8941-8956.
[25] Pu Yuting. Research on the Application of Data Mining in Computer
Network Virus Defense. 2021(2019-3):217-218.
[26] Pei Pei. Analysis on the Application of Visual Communication Design in
Packaging——Take Bamboo Leaf Green Tea Packaging Design as an
Example. Artwork Jian, 2020(33):59-60.
[27] Rizk M, Baghdadi A, Jezequel M, et al. No-instruction-set-computer
design experience of flexible and efficient architectures for digital
communication applications: two case studies on MIMO turbo detection
and universal turbo demapping. Design Automation for Embedded
Systems, 2021, 25(1):1-42.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the Creative
Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en_US
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2022.18.12
Xiaoyu Yan
E-ISSN: 2224-3488
95
Volume 18, 2022