Semantic and structural image segmentation for prosthetic vision
- PMID: 31995568
- PMCID: PMC6988941
- DOI: 10.1371/journal.pone.0227677
Semantic and structural image segmentation for prosthetic vision
Abstract
Prosthetic vision is being applied to partially recover the retinal stimulation of visually impaired people. However, the phosphenic images produced by the implants have very limited information bandwidth due to the poor resolution and lack of color or contrast. The ability of object recognition and scene understanding in real environments is severely restricted for prosthetic users. Computer vision can play a key role to overcome the limitations and to optimize the visual information in the prosthetic vision, improving the amount of information that is presented. We present a new approach to build a schematic representation of indoor environments for simulated phosphene images. The proposed method combines a variety of convolutional neural networks for extracting and conveying relevant information about the scene such as structural informative edges of the environment and silhouettes of segmented objects. Experiments were conducted with normal sighted subjects with a Simulated Prosthetic Vision system. The results show good accuracy for object recognition and room identification tasks for indoor scenes using the proposed approach, compared to other image processing methods.
Conflict of interest statement
The authors have declared that no competing interests exist.
Figures
![Fig 1](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/6988941/bin/pone.0227677.g001.gif)
![Fig 2](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/6988941/bin/pone.0227677.g002.gif)
![Fig 3](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/6988941/bin/pone.0227677.g003.gif)
![Fig 4](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/6988941/bin/pone.0227677.g004.gif)
![Fig 5](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/6988941/bin/pone.0227677.g005.gif)
![Fig 6](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/6988941/bin/pone.0227677.g006.gif)
![Fig 7](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/6988941/bin/pone.0227677.g007.gif)
![Fig 8](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/6988941/bin/pone.0227677.g008.gif)
![Fig 9](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/6988941/bin/pone.0227677.g009.gif)
![Fig 10](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/6988941/bin/pone.0227677.g010.gif)
![Fig 11](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/6988941/bin/pone.0227677.g011.gif)
![Fig 12](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/6988941/bin/pone.0227677.g012.gif)
Similar articles
-
PVGAN: a generative adversarial network for object simplification in prosthetic vision.J Neural Eng. 2022 Sep 7;19(5). doi: 10.1088/1741-2552/ac8acf. J Neural Eng. 2022. PMID: 35981530
-
Image processing strategies based on saliency segmentation for object recognition under simulated prosthetic vision.Artif Intell Med. 2018 Jan;84:64-78. doi: 10.1016/j.artmed.2017.11.001. Epub 2017 Nov 10. Artif Intell Med. 2018. PMID: 29129481
-
Image Processing Strategies Based on a Visual Saliency Model for Object Recognition Under Simulated Prosthetic Vision.Artif Organs. 2016 Jan;40(1):94-100. doi: 10.1111/aor.12498. Epub 2015 May 15. Artif Organs. 2016. PMID: 25981202
-
Simulating prosthetic vision: I. Visual models of phosphenes.Vision Res. 2009 Jun;49(12):1493-506. doi: 10.1016/j.visres.2009.02.003. Vision Res. 2009. PMID: 19504749 Review.
-
Rehabilitation regimes based upon psychophysical studies of prosthetic vision.J Neural Eng. 2009 Jun;6(3):035009. doi: 10.1088/1741-2560/6/3/035009. Epub 2009 May 20. J Neural Eng. 2009. PMID: 19458400 Review.
Cited by
-
Great expectations: Aligning visual prosthetic development with implantee needs.medRxiv [Preprint]. 2024 Mar 14:2024.03.12.24304186. doi: 10.1101/2024.03.12.24304186. medRxiv. 2024. PMID: 38559196 Free PMC article. Preprint.
-
Towards biologically plausible phosphene simulation for the differentiable optimization of visual cortical prostheses.Elife. 2024 Feb 22;13:e85812. doi: 10.7554/eLife.85812. Elife. 2024. PMID: 38386406 Free PMC article.
-
An actor-model framework for visual sensory encoding.Nat Commun. 2024 Jan 27;15(1):808. doi: 10.1038/s41467-024-45105-5. Nat Commun. 2024. PMID: 38280912 Free PMC article.
-
A novel simulation paradigm utilising MRI-derived phosphene maps for cortical prosthetic vision.J Neural Eng. 2023 Aug 10;20(4):046027. doi: 10.1088/1741-2552/aceca2. J Neural Eng. 2023. PMID: 37531948 Free PMC article.
-
Object recognition and localization enhancement in visual prostheses: a real-time mixed reality simulation.Biomed Eng Online. 2022 Dec 24;21(1):91. doi: 10.1186/s12938-022-01059-7. Biomed Eng Online. 2022. PMID: 36566183 Free PMC article.
References
-
- Zhou DD, Dorn JD, Greenberg RJ. The Argus® II retinal prosthesis system: An overview. In: 2013 IEEE International Conference on Multimedia and Expo Workshops (ICMEW). IEEE; 2013. p. 1–6.
-
- Lovell NH, Hallum LE, Chen S, Dokos S, Byrnes-Preston P, Green R, et al. Advances in retinal neuroprosthetics Handbook of Neural Engineering. 2007; p. 337–356.
Publication types
MeSH terms
Associated data
Grants and funding
LinkOut - more resources
Full Text Sources