Purchase this article with an account.
Yoko Mizokami, Yuki Takahashi, Hirohisa Yaguchi; Stable colorfulness perception of scene through haze. Journal of Vision 2016;16(12):216. doi: 10.1167/16.12.216.
Download citation file:
© 2017 Association for Research in Vision and Ophthalmology.
The saturation of visual scene decreases by haze due to environments (e.g. fog) and optical properties of eyes (e.g. aging of crystalline lens). However, we would be able to maintain stable colorfulness perception of a scene if we have a mechanism to compensate deceased saturation by haze. We previously examined the influence of haze on the colorfulness perception of natural images using foggy filters, and showed that it was almost the same with and without the foggy filters even immediately after wearing the filters, suggesting a strong instantaneous colorfulness compensation mechanism (ICVS2015). Here, we further examine if the strength of colorfulness compensation is influenced by the pattern of stimuli. We tested conditions viewing natural images through actual foggy filters as well as simulated images, using a colorfulness matching method. Observers memorized the colorfulness of a stimulus image without foggy filters. Then they adjusted the colorfulness of a natural image through foggy filters (or a simulated foggy image) to match the memorized colorfulness of the image by controlling the metric chroma of the image. We examined natural images and also simple patterns such as the array of squares with single or multiple colors. Each observer run three sessions under each condition. Results show that colorfulness settings of images through the filters was almost the same as those without the filters in all conditions, suggesting that we can adjust our perception to deceased saturation by haze and maintain stable colorfulness perception for any stimuli. The variance of results was smaller for the natural images compared to other simple patterns in general, implying that the compensation of deceased saturation is more stable for natural scenes with complex information.
Meeting abstract presented at VSS 2016
This PDF is available to Subscribers Only