Answers are grounded in this paper and cite their sources.
Portable image acquisition devices often operate in challenging outdoor environments. The quality of the acquired scenes is often low due to the low dynamics of current sensors. Moreover, these sensors are typically equipped with low-cost hardware. Hence, the image-enhancement algorithms are bounded to work in real time using limited resources and possibly require low power. We introduce a novel method to enhance the digital images acquired in critical lighting conditions, producing high-quality images for most applications. Our algorithm separates the image into illumination and reflectance , as in the ldquoRetinexrdquo theory. These two channels are separately processed since they give different contributions to the image. We propose a simple and fast operator to estimate the illumination component employing a spatially recursive rational filter. We show some results of an extensive experimental phase to show that to separately process the illumination and reflectance channels is advantageous since the image features are better exploited. A number of measures are computed to validate our choices.
No takes yet. Share an insight, caveat, or question.
Marsi et al. (2008) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: