Image processing strategies based on saliency segmentation for object recognition under simulated prosthetic vision
作者：
Highlights：
• We developed an optimized saliency segmentation method for automatically extracting the object of interest.
• Two image processing strategies were further presented for enhancing the object of interest.
• Both strategies improved object recognition performance in low-resolution prosthetic vision.
• Object recognition would be affected by object segmentation under simulated prosthetic vision.
• Under low-resolution prosthetic vision, the paired-interrelated objects had a positive impact on recognition.
摘要
•We developed an optimized saliency segmentation method for automatically extracting the object of interest.•Two image processing strategies were further presented for enhancing the object of interest.•Both strategies improved object recognition performance in low-resolution prosthetic vision.•Object recognition would be affected by object segmentation under simulated prosthetic vision.•Under low-resolution prosthetic vision, the paired-interrelated objects had a positive impact on recognition.
论文关键词：Visual prosthesis,Simulated prosthetic vision,Saliency segmentation,Image processing strategy,Objects recognition
论文评审过程：Received 9 October 2016, Revised 3 November 2017, Accepted 7 November 2017, Available online 10 November 2017, Version of Record 5 February 2018.
论文官网地址：https://doi.org/10.1016/j.artmed.2017.11.001
