Deep Learning-Based Machine Color Emotion Generation

Deep Learning-Based Machine Color Emotion Generation

Tongyao Nie, Xinguang Lv
DOI: 10.4018/IJMCMC.325349
Article PDF Download
Open access articles are freely available for download

Abstract

This paper investigates generating machine color emotion through deep learning. The grayscale image colorization model's training process resembles human memory color. Sixty images were recolored and quality evaluated to explore machine generated color impressions. Six experimental samples were recolored under D65, A, CWF, and TL84 light sources. Changes in lightness, chroma, and hue angle compared the original and colorized images, exploring light source effects on machine color perception. Analyzing differences in coloring results within the CIEL* a* b* color space for pixels with equal grayscale verified machine color emotion generation. Results show the machine learns to form color impressions from samples. Different light source color temperatures impact color prediction accuracy. The machine accurately colors images based on semantic context, demonstrating spontaneous color emotion generation through deep learning. This research positively contributes to the development of intelligent devices with color emotion.
Article Preview
Top

Introduction

Emotional intelligence is an advanced stage in the development of artificial intelligence (AI); it is at the forefront of contemporary information development, and a new product that incorporates emotion into the field of information science. The extended cognitive theory argues that with the development of computer and AI technologies, human cognitive activities must rely on intelligent devices to be completed. The cognitive subject not only is an individual natural person but also includes these electronic devices, which are cooperative or symbiotic (van Holland, 2013). It provides ideas that can be applied to developing AI and emotional machines. Developing emotional robots is a trend and requirement for changing weak AI to strong AI and super AI (Fjelland, 2020). The creation process has eight steps: discovering the problem, acquiring knowledge, gathering relevant information, conceiving ideas, generating ideas, combining ideas, selecting the best idea, and externalizing the concept. Today’s machines are not equipped with the ability to discover problems and formulate them, nor can the data selection and collection rules be set by the devices themselves. However, in data processing, data has an advantage that humans cannot match. In this case, the model of human-machine collaboration is still the optimal solution. Suppose the ability of emotion recognition and expression is given to machines. In that case, this advancement will not only push AI to a higher platform but also assist in human emotion understanding and expression. It will also facilitate the overall development of human beings (Korsmeyer, 1999).

Although previous researchers have described the creation of many robots and architectures for autonomous agents that can mimic human emotion, most of this research has focused on social domains, such as medical services (Esteva et al., 2017), teaching services (Ling, 2022; Yang, 2022), and family services (Breazeal, 2003; Lee-Johnson & Carnegie, 2010), mainly possessing speech recognition communication technologies (Lafaye et al., 2014).

In contrast, robots with color emotion have been less developed. There are few applications in the field of color perception and art creation. However, vision is the most potent way humans perceive objective things in communication with the outside world (Li et al., 2022). More than 87% of people obtain information visually, with 70% to 80% of this information obtained from color (Wu & Fang, 2022; Xu et al., 2022), so the emotion conveyed by color is essential (He & Lv, 2022). Color emotion refers to the light information of different frequencies of color acting on the human visual organs, through the optic nerve to the brain, and then associated with memories and experiences, thus forming a series of color psychological responses.

A colorimeter obtains the current machine perception of color by measuring the spectral properties of the material surface to obtain color data, which for color perception, remains only at the level of physical properties and color measurement. This measurement is formed by three links: light source, object, and sensor. However, color is a subjective response of the human visual system to electromagnetic radiation in the visible spectrum with wavelengths between 380 nm and 780 nm. Four factors are required to form human color perception: the light source, the object, the eye, and the brain. Human color perception involves physics, physiology, and psychology (Zeger et al., 2021); therefore, compared with humans, machines lack psychological perception of color, and their perception of color does not resonate with humans. This limitation hinders better human-machine communication and collaboration.

Complete Article List

Search this Journal:
Reset
Volume 15: 1 Issue (2024)
Volume 14: 1 Issue (2023)
Volume 13: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 12: 4 Issues (2021)
Volume 11: 4 Issues (2020)
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing