We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Determining the parameters of emotion by analyzing environmental images captured by a mobile device.
- Authors
Hwang, Hyeonsang; Lee, Eui Chul
- Abstract
This paper proposes a method to quantitatively extract human emotions by analyzing images of the surrounding environment captured by a smartphone camera in real time. In the field of psychology, it is known that visual elements such as colors and complexity affect human emotions. Based on this knowledge, we developed an application for the extraction of emotions in real time using the colors and spatial complexity of images obtained with Android smartphones. Among the color components of images, the hue component, which indicates the color, is extracted as the color element and the spatial complexity is extracted by quantitatively determining the high- and low-frequency components that are visible in the images. The corresponding two-dimensional emotion is identified by using two support vector regression modules. The results show that the root-mean-square error between the estimated emotion and the subjectively evaluated emotion is approximately 0.36 on the −1 to +1 axes plane.
- Subjects
ANDROID (Operating system); EMOTIONS; TIME management; IMAGE; COLOR
- Publication
Multimedia Tools & Applications, 2019, Vol 78, Issue 20, p28375
- ISSN
1380-7501
- Publication type
Article
- DOI
10.1007/s11042-017-5342-1