A Modified Structural Similarity Index With Low Computational Complexity

• Main Authors: Bong, David Liang Bong, Loh, Woei-Tan
• Format: Proceeding
• Language: English; English
• Published: 2019
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://ir.unimas.my/id/eprint/28483/5/A%20MODIFIED.pdf; http://ir.unimas.my/id/eprint/28483/3/ISPACS%202019_Program%20Book%20-%20Copy.pdf; http://ir.unimas.my/id/eprint/28483/; https://ieeexplore.ieee.org/xpl/conhome/1001126/all-proceedings

Structural Similarity Index (SSIM) has been a benchmark method for image quality assessment (IQA). This is due to its simplicity and good performance. In this paper, we propose a modified SSIM method that reduces the computational complexity with comparable performance. Instead of computing similarities on local windows, the proposed method computes global information similarities. The proposed method omits the luminance part similarities in SSIM due to its less crucial role in assessing image quality. From the presented results, the proposed method has a much lower computational time and comparable performance compared to SSIM.