Application of Deep Neural Networks and Human-Computer Interaction Technology in Art and Design Area

Abstract

Technological development has given a new dimension to the art and design area. This research implements the deep learning model for human-computer integration technology in art and design with wireless sensor networks. In this human-computer interaction, each person and the devices are treated as network nodes. The user will act as a node requesting access privileges for utilizing the calligraphy design to learn the strokes in the self-pace mode for studies. In this research, the human-computer interaction is implemented for recognizing the strokes and curves in the calligraphy images loaded by the user. This identification will aid the user in learning different styles of calligraphy. The proposed fruit-fly optimization algorithm (FFOA) method is analyzed for high classification accuracy, less delay time, and reduced noise parameters against specific existing algorithms.

This work is licensed under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.

1. Introduction

Technology that uses computer-aided arts and crafts art design to save workers from laborious and time-consuming tasks and obligations has recently been introduced. It is possible to improve the quality and efficiency of arts and crafts art products by following conventional arts and crafts art design concepts [1]. Because of artificial intelligence [2], it has become possible for computer-aided arts and crafts to climbing to previously inconceivable heights in terms of both quality and efficiency [3]. A network of merchants, distributors, transporters, storage facilities, and suppliers is required to create and distribute a product to a customer. In order to differentiate themselves from their competition, companies from a wide range of sectors work together. The raw materials used in production, how and when they are procured, and the whole supply chain are all taken into account. These raw ingredients are transformed into final goods via manufacturing [3]. A network of distributors, warehouses, and merchants is formed to ensure that these items reach their intended customers.

In the last two years, the use of electronic computers has increased at a pace of 40% each year, or every two years. With an interactive computer-aided design system, work duplication is reduced, the design process is accelerated, and creative ideas are enhanced. This may have a substantial influence on the content and quality of the finished product [4]. Many computer-generated artworks are published each year as a result of the rapid advancement in artificial intelligence technologies. On a flat surface, many arts and crafts have the quality of repeating themselves again and over again, as evidenced by numerous ancient artefacts [5]. Art and craft projects can be made on computer displays or conventional colour (or black-and-white) TVs using the fundamental units of graphic design, the overarching concept of composition, and the automated generating technology for interconnecting pictures. Metals (such as stainless steel), wood (such as oak), leather (such as calfskin), and other materials may be used to produce mosaic graphics that automatically interlock, replicate, and dry without drying the joint money graphic automatic generating technology [6].

It is possible to deliver location-based services at any time and from any location thanks to a distributed geographic information system in a mobile information system that uses high-performance wired and wireless communication networks (GIS). Integration of technologies such as GIS (geographical information systems), GPS (global positioning systems), mobile communication systems, and computer networks may be used to provide industrial and public information services among other things [7]. It is possible to utilise a GIS application technique to communicate and publish information about a user’s position using mobile devices such as mobile phones in a cost-effective manner. On the one hand, if you work in a field where data is extremely mobile and changes regularly (like commercial trade or government), a mobile information system might be a great asset. On the other hand, researchers in the field of artificial intelligence may now use their findings to create computer-aided art, which broadens both the scope and the number of applications for artificial intelligence [8]. In recent years, the field of artificial intelligence has grown to be quite active, and as a result, an increasing number of computer art pieces are becoming accessible each year [9]. A new technological field called artificial intelligence (AI) has emerged in the computer world, which analyses and develops ideas and methods, techniques, and applications that are meant to duplicate, extend, and expand human intellectual capabilities [10]. Artificial intelligence may be used to make computers more human-like, allowing them to be used in the most advanced manner possible. They have raised expectations for the capabilities of computers as the number of people who use them in everyday life grows [11]. Using artificial intelligence, computer technology has expanded to previously inconceivable heights of complexity. Artificial Intelligence (AI) presently has three main fields of study: intelligent interfaces, data mining, and topic and multiagent systems. With data mining, the main goal is to find particular individuals in a large and diverse collection of application data that is incomplete, unclear, and created carelessly. The research focuses on the ability of artificial intelligence to face an entity in the selection layer of belief, desire, intention, and ability, amongst other things, as the topic of the study [12]. Having important or possibly important knowledge is the first step in the process of knowing. Recent advances in the study of artificial intelligence, particularly in the domain of thinking abilities, have been notable. Fuzzy processing and parallel processing are also on the horizon. One of the most exciting new directions in technology is the development of artificial neural networks (ANNs). As artificial neural networks continue to improve, we should expect a significant improvement in society’s advancement. These computers were first suggested in the 1970s and have been a major part of the development process ever since then [13]. A great number of manual workers are freed up by computer-aided arts and crafts in the workplace, allowing them to devote more time and effort to the creation of new technologies, as well as the study and development of creative forms of expression. Art and craft production may be made more efficient by the application of computer-aided arts and crafts. Canned arts and crafts not only facilitate production, but they also remove a significant number of the challenges that artists confront while creating new works of art [14]. As a result of using computer-aided arts and crafts technology, young designers who lack substantial design experience may be able to rise to a higher design level, enhance their design grade, and minimise the great need for designers with extensive design experience. Computer-aided design (CAD) for the arts might gain greatly from artificial intelligence. CAD is the use of computers, which helps in the modification, optimization, and analysis of designs. Computer-aided design may be found in many fields of arts and crafts technology [15]. Due to the present conditions, some advanced analytical abilities can be provided by CAD. Computer-aided process design can only apply artificial intelligence in the early stages of development, and it is limited to intelligent thinking at this time. In practice, there are no distinct varieties of artificial intelligence [16]. Instead, a variety of applications are combined to create a more complete form [17].

CAPP is used in art and design problems which can be solved with artificial intelligence technology as computer-aided art design advances. CAPP stands for computer-aided process planning. It is a technology that can be used for making plans for a product or manufacturing one. It is a link between CAD and CAM in that it offers planning for the designing part. This is because artificial intelligence mimics people’s thinking patterns, allowing for the acquisition of knowledge and reasoning about them. When artificial intelligence (AI) is included in a CAPP, it is considered to be intelligent. For an intelligent CAPP system, there is a wide range of different intelligent methodologies, including fuzzy reasoning and chaotic theory [18]. Intelligence may be demonstrated by combining the advantages of artificial neural networks with perceptual image thinking in practical applications. Fuzzy inference methods can be used in conjunction with other forms of logical reasoning to enhance a person’s overall intellect and thinking capacity [19]. As compared to mechanical product design theory, there is very little research on CAPP’s design theory available. The foundation layer, the tool layer, and the platform layer of the design theory and methodology all serve various purposes. This overlap between CAPP theory and mechanical product design theories is particularly apparent in terms of the intelligent design process [20]. The three levels of intelligent CAPP design theory and method architecture are regarded as the foundation scientific layer, the information technology, and intelligent design methodology layers (or intelligent design method), respectively.

In the field of art and craft design, there is high applicability for development. There are many variables that can change at any point during the design and production process. In computer-aided art design, these characteristics are likely to cause complications, especially when combined [21]. Artificial neural networks have enhanced the design of computer-aided arts and crafts (CAPP) and efficiently tackled these problems. As a means of overcoming the problem of network connectivity, artificial neural networks (ANNs) based on biological principles are being created. Many nonlinear processing units identified have been found here. Modern information technology units are very comparable to those in today’s technical environment [22]. Associative memory, self-organization, and self-learning are only a few properties of the artificial neural network that may be utilized to solve issues when given a task to tackle. In computer-aided art design, artificial neural network technology is the most advantageous since it is able to adapt and adjust the system as necessary. While keeping track of its flexibility, the neural network will also be able to examine its beginning core values and real system conditions [23]. The special monitoring approach consists of deduction and adjustment at each back node and pushing and adjusting at the end of the neural network. When it comes to product creativity and craftsmanship, there is an abundance of ambiguity that can only be resolved via experience. Developing a search or derivative system using a CAPP system that depends solely on group technology would fail since it would not be able to collect different types of information effectively [24]. The CAPP system uses artificial neural network technology as one of the most important AI applications. Artificial neural networks are based on the same principles as organic nerve systems when it comes to coping with the current environment. This system’s key characteristic is that it is made up of a large number of tiny nonlinear processing units that are linked in parallel and stores information in a dispersed manner [25]. This study focused on evaluating human-computer interaction technology in art and design using deep neural networks.

The contributions of this study are as follows:

(1) It focused on analysing the image data that was either transferred to the user or made only for the display with effective visualization

(2) It analyses the accurate validation of the recognised image by using a deep learning algorithm

(3) It uses equations of motion for categorising human targets and objects

(4) The intelligent system is used for the identification of the artwork and classification according to the required category

2. Materials and Methods

The dataset utilized in this research is the collection of Chinese character writings by famous Chinese calligraphers. This collection includes approximately 5,350 JPEG images which are of 64 × 64 pixels. The collected images are validated and transformed into the standardized shape by the person who prepared them. For training and testing the image, the dataset is divided into 80% and 20%, respectively. In this research, the recognised image data are either transferred to the user or made only the display with a better visualisation.

2.1. Motivation of the Study

Deep learning is an essential component of human-computer interaction (HCI) for art and design processes through interactive control systems with the aid of wireless sensor networks. In general, human-computer interaction is also dependent on the flexibility of a visual analytics system. It can be used in HCI to encapsulate non-invasive human hand art by directly stimulating just the human hand. Accurate validation and analysis of human art and design through real-time splash damage instances is considered difficult in HCI due to the complex background. Moreover, human hand art is highly dynamic, so addressing the noise problem requires a flexible approach.

Furthermore, structural thermal imaging was created to overcome the traditional transmitter problem, including face recognition and human identification in single target scenarios. Destination mobility and body movement investigate human art and design behaviour in spatial thermal imaging. Equations of motion are used to classify human targets and objects. The intelligent system plays a major role in the identification of the artwork and classification thereof according to the required category. This research work focuses on the classification of art data based on the order design optimization technique (Figure 1).

[figure omitted; refer to PDF]

Table 1

Performance result for the evaluation of human-computer interaction technology in art and design by using mean square error.

Total number of datasets in different methods	Mean square error overall accuracy (%)
LDA-ELM	47
MUST-BSSA	45
HMM-SVD	25
FFOA	49

The simulation results indicate that, when compared with existing approaches, the suggested technique could evaluate the human target angle with high precision. Figure 3 depicts the average recognition ratio achieved with our suggested FFOA methods. From the figure, it is observed that the proposed FFOA algorithm shows the highest recognition ratio of the highly needed user with a minimum recognition ratio of about 85%. Also, the percentage is fluctuating between 85 and 90% even when there is an increase in the number of records or the user’s request. The other existing algorithms used in this research work LDA-ELM, MUST-BSSA, and HMM-SVD have achieved a maximum accuracy of 79%, 83%, and 79%, respectively. The fluctuating accuracy ratios are 45% to 83%, 45% to 83%, and 35% to 79%, respectively. Also, the existing algorithm’s accuracy faces a heavy downfall when there is an increase in the number of users.

[figure omitted; refer to PDF]

Figure 4 depicts the suggested FFOA method’s delay time. The ratio of a determined design to undesirable environmental noise components demonstrates the effectiveness of an art and design measurement. A transmitter design provides more information to predict intension, increasing predictive performance. Nonetheless, noise sources from different sources are feasible, and art and design analysis may well be contaminated. The amplifier is designed and then used to dismiss or eliminate noise levels to maximize the transmission ratio. Though the number of datasets increases, the delay in making responses is significantly less than the existing algorithms, with a minimum difference of 45% lesser duration than the existing algorithm. The minimum delay in data transfer observed by the existing LDA-ELM, MUST-BSSA, and HMM-SVD algorithms is 44%, 45%, and 46%, respectively, whereas the proposed FFOA has shown its least by 25% (see Table 2).

[figure omitted; refer to PDF]

The existing algorithms considered for performance analysis in this research are as follows: motor unit spike trains with blind source separation algorithm (MUST-BSSA), hidden Markov model and singular value decomposition (HMM-SVD) methods, and the linear discriminant analysis and extreme learning machine (LDA-ELM) method. Compared to these models, the proposed optimization method for noninvasive human-computer interaction named “fruit-fly optimization algorithm (FFOA)” outperforms them with high classification accuracy, less delay time, and reduced noise. Table 3 represents this comparison analysis and shows that as the dataset increases, the error rate in the proposed system is greatly reduced to a minimum of 19%, which is the least among other existing algorithms. Its percentage is 13%, 16%, and 19% less when compared with the HMM-SVD, MUST-BSSA, and LDA-ELM methods, respectively.

Table 3

Comparison result for human-computer interaction technology in art and design.

Number of data	LDA-ELM (%)	MUST-BSSA (%)	HMM-SVD (%)	FFOA (%)
10	80	83	79	45
20	78	75	72	43
30	65	79	65	39
40	45	64	60	35
50	59	55	54	33
60	62	52	49	29
70	49	48	45	31
80	43	42	39	26
90	40	35	33	21
100	38	35	32	19

4. Conclusions

The development of artificial intelligence paved the way for designing computer-aided arts and crafts and enhanced the quality of production. The enhancement of artificial technology and different algorithms and technologies makes the design of computer-aided arts and crafts easier. Deep learning techniques place a high demand on the computing power of experimental equipment and computers. This study focused on evaluating human-computer interaction technology in art and design using artificial intelligence. The proposed fruit-fly optimization algorithm (FFOA) method is analysed for high classification accuracy, less delay time, and reduced noise parameters against specific existing algorithms. It is highly suggested to enhance the accuracy and complexity of the model for future studies.

Acknowledgments

The study was supported by Important level scientific research project of Anhui Vocational and Technical College of Commerce and Trade (No. 2020KYR15), Platform base project of Anhui Business Vocational and Technical College (No. 2020zdy07), and Anhui Philosophical Planning Project “inclusive”-Research on the artistic style of tea packaging in the Republic of China (No. AHSKQ2021D133).

References

[1] J. Rahm-Skågeby, L. Rahm, "HCI and deep time: toward deep time design thinking," Human-Computer Interaction,DOI: 10.1080/07370024.2021.1902328, 2021.

[2] K. Browne, B. Swift, H. Gardner, "Critical challenges for the visual representation of deep neural networks," Human and Machine Learning, pp. 119-136, DOI: 10.1007/978-3-319-90403-0_7, 2018.

[3] P. Heracleous, Y. Mohammad, A. Yoneyama, "Deep convolutional neural networks for feature extraction in speech emotion recognition human-computer interaction," Recognition and Interaction Technologies, pp. 117-132, DOI: 10.1007/978-3-030-22643-5_9, 2019.

[4] A. Thanda, S. M. Venkatesan, "Audio visual speech recognition using deep recurrent neural networks," Lecture Notes in Computer Science, pp. 98-109, DOI: 10.1007/978-3-319-59259-6_9, 2017.

[5] H. Seo, G. Luo, "Generating 3D facial expressions with recurrent neural networks," Intelligent Scene Modeling and Human-Computer Interaction, pp. 181-196, DOI: 10.1007/978-3-030-71002-6_11, 2021.

[6] P. Deshpande, "On potential application of interaction design for placemaking," Human-Computer Interaction Series, pp. 321-343, DOI: 10.1007/978-3-319-30028-3_15, 2016.

[7] E. Baka, N. Magnenat Thalmann, "Human-technology interaction: the state-of-the-art and the lack of naturalism," Intelligent Scene Modeling and Human-Computer Interaction, pp. 221-239, DOI: 10.1007/978-3-030-71002-6_13, 2021.

[8] C. Troussas, A. Krouska, M. Virvou, "Adaptive e-learning interactions using dynamic clustering of learners’ characteristics," Proceedings of the 2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA),DOI: 10.1109/iisa.2019.8900722, .

[9] Y. Ji, Y. Yang, F. Shen, H. T. Shen, X. Li, "A survey of human action analysis in HRI applications," IEEE Transactions on Circuits and Systems for Video Technology, vol. 30 no. 7, pp. 2114-2128, DOI: 10.1109/tcsvt.2019.2912988, 2020.

[10] T. Crocken, A. Washington, "Strategies and tools for digital repository selection and migration," Proceedings of the 2019 ACM/IEEE Joint Conference on Digital Libraries (JCDL), pp. 446-447, DOI: 10.1109/jcdl.2019.00112, .

[11] S. Amara, R. R. Subramanian, "Collaborating personalized recommender system and content-based recommender system using TextCorpus," Proceedings of the 2020 6th International Conference on Advanced Computing and Communication Systems (ICACCS), pp. 105-109, DOI: 10.1109/icaccs48705.2020.9074360, .

[12] B. Guan, L. Hu, P. Liu, H. Xu, Z. Fu, Q. Wang, "dpSmart: a flexible group based recommendation framework for digital repository systems," Proceedings of the 2019 IEEE International Congress on Big Data (BigDataCongress), pp. 111-120, DOI: 10.1109/bigdatacongress.2019.00028, .

[13] T. Hassan, P. A. Shah, F. A. Khan, "Fusion of motivational strategies with recommender system innovative framework for digital repositories," Proceedings of the 2016 International Conference on Computing, Electronic and Electrical Engineering (ICE Cube), pp. 32-36, DOI: 10.1109/icecube.2016.7495249, .

[14] P. Jomsri, "FUCL mining technique for book recommender system in library service," Procedia Manufacturing, vol. 22, pp. 550-557, DOI: 10.1016/j.promfg.2018.03.081, 2018.

[15] P. Knoth, L. Anastasiou, A. Charalampous, Towards Effective Research Recommender Systems for Repositories, pp. 26-30, 2017.

[16] M. Borovič, M. Ferme, J. Brezovnik, S. Majninger, K. Kac, M. Ojsteršek, "Document recommendations and feedback collection analysis within the slovenian open-access infrastructure," Information, vol. 11 no. 11,DOI: 10.3390/info11110497, 2020.

[17] M. Gupta, A. Thakkar, Aashish, V. Gupta, D. P. S. Rathore, "Movie recommender system using collaborative filtering," Proceedings of the 2020 International Conference on Electronics and Sustainable Communication Systems (ICESC), pp. 415-420, DOI: 10.1109/icesc48915.2020.9155879, .

[18] R. Singla, S. Gupta, A. Gupta, D. K. Vishwakarma, "FLEX: a content based movie recommender," Proceedings of the 2020 International Conference for Emerging Technology (INCET),DOI: 10.1109/incet49848.2020.9154163, .

[19] K. Dahdouh, L. Oughdir, A. Dakkak, A. Ibriz, "Smart courses recommender system for online learning platform," Proceedings of the 2018 IEEE 5th International Congress on Information Science and Technology (CiSt), pp. 328-333, DOI: 10.1109/cist.2018.8596516, .

[20] T. Wongkhamchan, A. Namvong, C. Surawanitkun, "Personalized recommender system using a social network based collaborative filtering technique," Proceedings of the 2019 16th IEEE International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON), pp. 846-849, DOI: 10.1109/ecti-con47248.2019.8955422, .

[21] A. Mughaid, I. Obeidat, B. Hawashin, S. AlZu’bi, D. Aqel, "A smart geo-location job recommender system based on social media posts," Proceedings of the 2019 Sixth IEEE International Conference on Social Networks Analysis, Management and Security (SNAMS), pp. 505-510, DOI: 10.1109/snams.2019.8931854, .

[22] X. Yang, M. Dong, X. Chen, K. Ota, "Recommender system-based diffusion inferring for open social networks," IEEE Transactions on Computational Social Systems, vol. 7 no. 1, pp. 24-34, DOI: 10.1109/tcss.2019.2950139, 2020.

[23] B. Liang, B. Xu, X. Wu, "A community-based collaborative filtering method for social recommender systems," Proceedings of the 2019 IEEE International Conference on Web Services (ICWS), pp. 159-162, DOI: 10.1109/icws.2019.00036, .

[24] S.-M. Choi, K. Jang, T.-D. Lee, A. Khreishah, W. Noh, "Alleviating item-side cold-start problems in recommender systems using weak supervision," IEEE Access, vol. 8, pp. 167747-167756, DOI: 10.1109/access.2020.3019464, 2020.

[25] C. Troussas, A. Krouska, C. Sgouropoulou, "Improving learner-computer interaction through intelligent learning material delivery using instructional design modeling," Entropy, vol. 23 no. 6,DOI: 10.3390/e23060668, 2021.