行動裝置的螢幕屬於高功耗硬體，用電量佔比高。在使用行動裝置時，螢幕開啓的頻率極高，例如撰寫電子訊息時、觀看影片時，均會保持螢幕開啓，在前述情境中螢幕用電平均達裝置總用電的15%以上。為了降低螢幕用電量，研究者提出局部降低亮度、關閉螢幕像素的發光通道，以及節能色彩的轉換等。前述各項方法均修改了呈現畫面，因此可能對使用者體驗造成負面影響，為了能在達到省電效果的同時，也兼顧使用者體驗，本研究提出一個依據使用者視覺注意力模型，對應用程式介面元件逐一進行節能色彩轉換的方法。本方法將整個應用程式介面拆分為各個介面元件，對各個元件分別採取不同的色彩轉換策略，並因應視覺注意力的分佈變化持續重新安排色彩，直至能達成省電效果，同時將視覺注意力變化保持在設定好的範圍區間，維持良好的使用者體驗。本系統適用於長時間停留於單一畫面的應用程式中，經過實驗驗證，本方法於預設設定下平均省電比率範圍為12%~35%，並觀察發現視覺顯著性變化和省電效果間存在關聯性，依據此關聯性提供設計省電色彩轉換規則的建議。

Display module is a part to be reckoned with when it comes to power consumption of mobile devices. Display modules usually consume more than 15% of the total power consumption while messaging in application, retouching photos, or playing videos. Therefore, researchers were dedicated to making power-saving mode based on screen features, such as reducing luminance of partial regions, shutting off subpixels, and doing color transformation. Each of the power-saving methods might affect the display effect, resulting in negative user experience. To deal with this problem, we present a method which can make mobile applications more energy-efficient by transforming the colors of the user interface in a bottom-up way, which divides the interface into every element on app layout and decides the color of each of them. This method also keeps user experience by using the feature of human visual salience in the phase of color transformation. With the method, we can rearrange the color of each app element on app layout to be more power-optimized, and simultaneously keep the user attention on modified app layout similar to the original app layout by doing saliency prediction analysis. With OLED mobile devices, this system is able to reduce, on average, between 12% and 35% of the display power consumption when using static mobile applications, which are usually stay on unchanging view. We find that the correlation between the changes of visual saliency distribution and the power saving effect, and base on the correlation, we provide some suggestions for designing power-saving color transformation rules.

[1] X. Chen, Y. Chen, Z. Ma, and F. C. A. Fernandes, "How is energy consumed in smartphone display applications?," presented at the Proceedings of the 14th Workshop on Mobile Computing Systems and Applications, Jekyll Island, Georgia, pp. 1-6, 2013.
[2] D. Shin, Y. Kim, N. Chang, and M. Pedram, "Dynamic voltage scaling of OLED displays," presented at the Proceedings of the 48th Design Automation Conference, San Diego, California , pp. 53-58, 2011.
[3] M. Park and M. Song, "Saving Power in Video Playback on OLED Displaysby Acceptable Changes to Perceived Brightness," Journal of Display Technology, vol. 12, no. 5, pp. 483-490, 2016.
[4] M. Dong and L. Zhong, "Chameleon: A Color-Adaptive Web Browser for Mobile OLED Displays," IEEE Transactions on Mobile Computing, vol. 11, no. 5, pp. 724-738, 2012.
[5] C.-H. Lin, C.-K. Kang, and P.-C. Hsiu, "CURA: A Framework for Quality-Retaining Power Saving on Mobile OLED Displays," ACM Transactions on Embedded Computing Systems (TECS), vol. 15, no. 4, pp. 1-25, 2016.
[6] D. Saha, A. Mandal, and S. Pal, "User Interface Design Issues for Easy and Efficient Human Computer Interaction: An Explanatory Approach," International Journal of Computer Sciences and Engineering, vol. 3, pp. 127-135, 2015.
[7] L. Bollini, "Beautiful interfaces. From user experience to user interface design," The Design Journal, vol. 20, no. sup1, pp. S89-S101, 2017.
[8] L. A. Leiva, Y. Xue, A. Bansal, H. R. Tavakoli, T. Köroðlu, J. Du, N. R. Dayama, and A. Oulasvirta, "Understanding Visual Saliency in Mobile User Interfaces," presented at the Proceedings of 22nd International Conference on Human-Computer Interaction with Mobile Devices and Services, Oldenburg, Germany, pp. 1-12, 2020.
[9] R. Valenti, N. Sebe, and T. Gevers, "Image saliency by isocentric curvedness and color," presented at the Proceedings of IEEE 12th International Conference on Computer Vision, pp. 2185-2192, 2009.
[10] P. Gupta, S. Gupta, A. Jayagopal, S. Pal, and R. Sinha, "Saliency Prediction for Mobile User Interfaces," presented at the Proceedings of IEEE Winter Conference on Applications of Computer Vision (WACV), pp. 1529-1538, 2018.
[11] K. W. Tan, T. Okoshi, A. Misra, and R. K. Balan, "FOCUS: a usable & effective approach to OLED display power management," presented at the Proceedings of the 2013 ACM international joint conference on Pervasive and ubiquitous computing, Zurich, Switzerland, pp. 573-582, 2013.
[12] M. Cornia, L. Baraldi, G. Serra, and R. Cucchiara, "A deep multi-level network for saliency prediction," presented at the Proceedings of 23rd International Conference on Pattern Recognition (ICPR), pp. 3488-3493, 2016.
[13] L. Itti, C. Koch, and E. Niebur, "A model of saliency-based visual attention for rapid scene analysis," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 20, no. 11, pp. 1254-1259, 1998.
[14] Z. Bylinskii, T. Judd, A. Oliva, A. Torralba, and F. Durand, "What Do Different Evaluation Metrics Tell Us About Saliency Models?," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 41, no. 3, pp. 740-757, 2019.
[15] N. Riche, M. Duvinage, M. Mancas, B. Gosselin, and T. Dutoit, "Saliency and Human Fixations: State-of-the-Art and Study of Comparison Metrics," in 2013 IEEE International Conference on Computer Vision, 2013, pp. 1153-1160.
[16] L. Punchoojit and N. Hongwarittorrn, "Usability Studies on Mobile User Interface Design Patterns: A Systematic Literature Review," Advances in Human-Computer Interaction, p. 6787504, 2017.
[17] G. Singh. "ComposeCookBook [Source code]." https://github.com/Gurupreet/ComposeCookBook (accessed 3/22, 2021).
[18] Google Inc. "Profile battery usage with Batterystats and Battery Historian." https://developer.android.com/topic/performance/power/setup-battery-historian (accessed 07/17, 2021).
[19] A. Carroll and G. Heiser, "An analysis of power consumption in a smartphone," presented at the Proceedings of the 2010 USENIX conference on USENIX annual technical conference, Boston, MA, pp. 21, 2010.