In this study, three salient region detection (SRD) methods and ten BISA models are jointly explored, and the output saliency maps from SRD methods are re-organized as the input of BISA models.
At last, experiments are conducted on three Gaussian blurring image databases and the prediction performance is evaluated.
We have reported experimental results on different IQA datasets including TID2013, LIVE, CSIQ. Note: You need to download the corresponding datasets.
- Saliency_Region_Detection --- The generation of salient region masks.
- sharpnessBISA --- blind image sharpness assessment methods
- main.py --- You need to change the base_Path value according to the location of the files.
- Matlab R2018a to implement main.m
- Python 3.6 to generate salient region mask of SRD_SORBD
- Windows operating system to execute the EXE files of SRD_SRIS and SRD_DPLSG
- A. Joshi, M. S. Khan, S. Soomro, A. Niaz, B. S. Han and K. N. Choi, "SRIS: Saliency-Based Region Detection and Image Segmentation of COVID-19 Infected Cases," IEEE Access, vol. 8, pp. 190487-190503, 2020.
- Zhou L, Yang Z, Zhou Z, et al. Salient Region Detection using Diffusion Process on a 2-Layer Sparse Graph[J]. IEEE Transactions on Image Processing, 2017, 26(12): 5882 - 5894.
- Zhu, Wangjiang and Liang, Shuang and Wei, Yichen and Sun, Jian. Saliency optimization from robust background detection.Proceedings of the IEEE conference on computer vision and pattern recognition 2014, 2814–2821.
- Dai G , Wang Z , Li Y , et al. Evaluation of no-reference models to assess image sharpness[C]// 2017 IEEE International Conference on Information and Automation (ICIA). IEEE, 2017.
If this idea is helpful, please cite this paper,
- Liu S, Yu S, Zhao Y, Tao Z, Yu H, Jin L. Salient region guided blind image sharpness assessment. Sensors. 2021 Jun 8;21(12):3963.