Eliminating Temporal Illumination Variations in Whisk-broom Hyperspectral Imaging

Abstract

We propose a method for eliminating the temporal illumination variations in whisk-broom (point-scan) hyperspectral imaging. Whisk-broom scanning is useful for acquiring a spatial measurement using a pixel-based hyperspectral sensor. However, when it is applied to outdoor cultural heritages, temporal illumination variations become an issue due to the lengthy measurement time. As a result, the incoming illumination spectra vary across the measured image locations because different locations are measured at different times. To overcome this problem, in addition to the standard raster scan, we propose an additional perpendicular scan that traverses the raster scan. We show that this additional scan allows us to infer the illumination variations over the raster scan. Furthermore, the sparse structure in the illumination spectrum is exploited to robustly eliminate these variations. We quantitatively show that a hyperspectral image captured under sunlight is indeed affected by temporal illumination variations, that a Naïve mitigation method suffers from severe artifacts, and that the proposed method can robustly eliminate the illumination variations. Finally, we demonstrate the usefulness of the proposed method by capturing historic stained-glass windows of a French cathedral.

Direct Alignment Of Narrow Field-Of-View Hyperspectral Data And Full-View Rgb Image

“…Hyperspectral to Gray Hyperspectral images offer a much sharper capture of the spectrum (high spectral resolution) than RGB images, leading to thousands of channels [4], i.e. N ≫ 3.…”

“…This system was set on a tripod to keep a static base during the scanning. The inevitable variation of natural illumination during such a time is eliminated thanks to a quick extra scanned line [4]. In the experiments, we've selected N = 663 channels (among 2068) of the spectrum ranging from 400nm to 700nm approximately corresponding to the visible light also captured by the RGB full-view camera.…”

“…Despite the long scanning time of whisk-broom HSI (i.e. one hour for 200×400 pixels of 2068 channels [4]), recent illumination variation compensation techniques [4] could avoid the usual need for light calibration [5]. This enables HSI's applicability to measurement of historic buildings and artifacts with natural lighting, particularly for challenging subjects, such as stainedglass windows [6] (Fig.…”

Comparison of Hyperspectral Imaging and Fiber-Optic Reflectance Spectroscopy for Reflectance and Transmittance Measurements of Colored Glass

“…This information allows for understanding the mechanisms responsible for the color of the glass, which can provide important insights into the technological development in glass production during a particular historical period or geographical region. The use of hyperspectral imaging, on the other hand, is quite limited, and very few works have been published so far on the topic, especially in comparison to the literature available for other typologies of artworks, such as painting and manuscripts [24][25][26][27]. Since stained glass is transparent, one of the main practical challenges is to have access to setups that can measure in transmittance.…”

“…In addition, the setup should be easily, as stained glass windows are usually located in buildings and can be difficult to access without scaffoldings or other supports. The easiest solution, and the most widely used [25][26][27], is to perform the acquisition by exploiting solar radiation as light source. However, variations of light distribution throughout the day, as well as the presence of external structures (vegetation, buildings, etc.)…”