Image Quality Assessment for Realistic Zoom Photos
Author:
Han Zongxi1ORCID, Liu Yutao2, Xie Rong1, Zhai Guangtao1ORCID
Affiliation:
1. Institute of Image Communication and Information Processing, Shanghai Jiao Tong University, Shanghai 200240, China 2. School of Computer Science and Technology, Ocean University of China, Qingdao 266100, China
Abstract
New CMOS imaging sensor (CIS) techniques in smartphones have helped user-generated content dominate our lives over traditional DSLRs. However, tiny sensor sizes and fixed focal lengths also lead to more grainy details, especially for zoom photos. Moreover, multi-frame stacking and post-sharpening algorithms would produce zigzag textures and over-sharpened appearances, for which traditional image-quality metrics may over-estimate. To solve this problem, a real-world zoom photo database is first constructed in this paper, which includes 900 tele-photos from 20 different mobile sensors and ISPs. Then we propose a novel no-reference zoom quality metric which incorporates the traditional estimation of sharpness and the concept of image naturalness. More specifically, for the measurement of image sharpness, we are the first to combine the total energy of the predicted gradient image with the entropy of the residual term under the framework of free-energy theory. To further compensate for the influence of over-sharpening effect and other artifacts, a set of model parameters of mean subtracted contrast normalized (MSCN) coefficients are utilized as the natural statistics representatives. Finally, these two measures are combined linearly. Experimental results on the zoom photo database demonstrate that our quality metric can achieve SROCC and PLCC over 0.91, while the performance of single sharpness or naturalness index is around 0.85. Moreover, compared with the best tested general-purpose and sharpness models, our zoom metric outperforms them by 0.072 and 0.064 in SROCC, respectively.
Subject
Electrical and Electronic Engineering,Biochemistry,Instrumentation,Atomic and Molecular Physics, and Optics,Analytical Chemistry
Reference80 articles.
1. Nakazawa, K., Yamamoto, J., Mori, S., Okamoto, S., Shimizu, A., Baba, K., Fujii, N., Uehara, M., Hiramatsu, K., and Kumano, H. (2021, January 11–16). 3D Sequential Process Integration for CMOS Image Sensor. Proceedings of the 2021 IEEE International Electron Devices Meeting (IEDM), San Francisco, CA, USA. 2. Zaitsu, K., Matsumoto, A., Nishida, M., Tanaka, Y., Yamashita, H., Satake, Y., Watanabe, T., Araki, K., Nei, N., and Nakazawa, K. (2022, January 12–17). A 2-Layer Transistor Pixel Stacked CMOS Image Sensor with Oxide-Based Full Trench Isolation for Large Full Well Capacity and High Quantum Efficiency. Proceedings of the 2022 VLSI Technology and Circuits, Honolulu, HI, USA. 3. Venezia, V.C., Hsiung, A.C.W., Yang, W.Z., Zhang, Y., Zhao, C., Lin, Z., and Grant, L.A. (2018). Second Generation Small Pixel Technology Using Hybrid Bond Stacking. Sensors, 18. 4. Yun, J., Lee, S., Cha, S., Kim, J., Lee, J., Kim, H., Lee, E., Kim, S., Hong, S., and Kim, H. (2022, January 12–17). A 0.6 μm Small Pixel for High Resolution CMOS Image Sensor with Full Well Capacity of 10,000e- by Dual Vertical Transfer Gate Technology. Proceedings of the 2022 VLSI Technology and Circuits, Honolulu, HI, USA. 5. Simulation-based study on characteristics of dual vertical transfer gates in sub-micron pixels for CMOS image sensors;Lee;Solid State Electron.,2022
|
|