Semantic Understanding and Task-Oriented for Image Assessment
This study focuses on image perception issues based on humans’ visual assessment. Semantic understanding and task-oriented are also considered in image assessment. The brightness and colorfulness attributes are selected to be the image assessment tasks in the study. The Linear Regression (LR) analysis and Non-Linear Regression (NLR) analysis methods are used to establish the image assessment models, which also compares their prediction ability. The visual assessment experiment was comprised of 90 participants. Four images were selected from the ISO standard by the focus group. The results showed that “brightness” and “colorfulness” remained stable in the predictive models of the LR and NLR methods. The results also demonstrated very high prediction ability in brightness and colorfulness in both the linear and non-linear models. The brightness attribute directly relates to the image’s lightness, and the colorfulness attribute directly relates to the image’s saturation. The simple semantic understanding and the single task oriented that assessed the brightness and colorfulness of the images are also very important in the image assessment experiment.
KeywordsBrightness Saturation Visual assessment Linear regression
The authors would like to thanks TTLA (Taiwan TFT LCD Association) for supporting this research and providing insightful comments. The authors would also like to thanks all observers for helpful in the experiments.
- 1.Newell, A.: Unified Theories of Cognition. Harvard University Press, Cambridge (1990)Google Scholar
- 2.Maeder, A.J., Eckert, M.: Medical image compression: quality and performance issues. In: Pham, B., Braun, M., Maeder, A.J, Eckert, M.P. (eds.) New Approaches in Medical Image Analysis, Proceedings of SPIE, vol. 3747, pp. 93–101 (1999)Google Scholar
- 4.Tsai, C.M., Guan, S.S., Juan, L.Y.G, Lai, Y.Y.: The scale on different physical attributes of images. In: 11th Congress of the International Colour Association on Proceedings, Sydney, Australia (2009)Google Scholar
- 8.Kurita, T., Saito, A.: A characteristic of the temporal integrator in the eye-tracing integration model of the visual system on the perception of displayed moving images. In: IDW 2002 Conference VHF2-1 on Proceedings, pp. 1279–128 (2002)Google Scholar
- 9.Chalmers, A.N.: Colour difference and colour preference in video imaging. In: 8th Congress of the International Colour Association on Proceedings, Kyoto, Japan, pp. 634–637 (1997)Google Scholar
- 11.Watson, A.B., Malo, J.: Video quality measures based on the standard spatial observer. In: IEEE ICIP, pp. 41–44 (2002)Google Scholar
- 13.Nguyen, A., Chandran, V., Sridharan, S.: Gaze trackign for region of interest coding in JPEG 2000. Sig. Process. Image Commun. 21, 356–377 (2006)Google Scholar
- 14.Civanlar, M. R.: Content adaptive video coding and transport. In Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, pp. 28–30 (2004)Google Scholar
- 15.Oda, K., Yuuki, A., Teragaki, T.: Evaluation of moving picture quality using the pursuit camera system. Euro Display 6(3), 115–118 (2002)Google Scholar
- 18.Choi, S.Y., Luo, M.R., Pointer, M.R., Rhodes, P.A.: Investigation of large display color image appearance i: important factors affecting perceived quality. J. Imaging Sci. Technol. 52(4), 040904-1–040904-11 (2008)Google Scholar
- 19.Choi, S.Y., Luo, M.R., Pointer, M.R., Rhodes, P.A.: Investigation of large display color image appearance ii: the influence of surround conditions. J. Imaging Sci. Technol. 52(4), 040905-1–040905-9 (2008)Google Scholar
- 21.Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E., Tatham, R.L.: Multivariate Data Analysis, 6th edn. Prentice Hall, Englewood Cliffs (2006)Google Scholar