No abstract

An improved spectral reflectance estimation method is developed to transform raw camera RGB responses to spectral reflectance. The novelty of our method is to apply a local weighted linear regression model for spectral reflectance estimation and construct the weighting matrix using a Gaussian function in CIELAB uniform color space. The proposed method was tested using both a standard color chart and a set of textile samples, with a digital RGB camera and by ten times ten-fold cross-validation. The results demonstrate that our method gives the best accuracy in estimating both the spectral reflectance and the colorimetric values in comparison with existing methods. IntroductionDigital cameras can be used to provide spectral data for many applications and thus the development of algorithms to calculate these spectral data from image RGB data is of prime importance. Different digital camera-based spectral imaging systems have been developed for practical applications, such as a camera with bandpass filters [1-3], a camera utilizing multiple illuminants [1,4,5], and a camera with three-channel responses (but a single RGB image) under a specific illuminant [6-13]. The latter system, which can be used to estimate the spectral reflectance from a single RGB image, has received considerably more interest because of the low cost of the devices with high resolution and their convenience in practical applications. The inherent problem of image registration which exists in optical bandpass filter-based spectral imaging systems [1][2][3]14] can be overcome by just one exposure using a conventional digital RGB camera. Accurate spectral and colorimetric estimation are both critical for practical applications and, since the digital RGB values are readily available from an image, an algorithm that accurately estimates the spectral reflectance from these RGB camera responses can be very useful.Methods to derive spectral data from camera responses can be divided into two types: model-based and training-based [15]. Since it is complex and expensive to characterize the camera sensitivity functions for the model-based method, the more easily implemented training-based method is both more convenient and more practical. Many training-based spectral estimation methods have been proposed in recent years [5][6][7][8][9][10][11][12][13]16]. Connah [6], Heikkinen [5], and Shen [8] proposed the use of a nonlinear regression method based on a polynomial model for spectral estimation, with consideration being given to the potential over-fitting problem in the polynomial-based regression model. Xiao et al. [10] illustrated that it is effective to combine the polynomial model with the eigenvector space of principal

A sequential weighted nonlinear regression technique from digital camera responses is proposed for spectral reflectance estimation. The method consists of two stages taking colorimetric and spectral errors between training set and target set into accounts successively. Based on polynomial expansion model, local optimal training samples are adaptively employed to recover spectral reflectance as accurately as possible. The performance of the method is compared with several existing methods in the cases of simulated camera responses under three kinds of noise levels and practical camera responses under the self as well as cross test conditions. Results show that the proposed method is able to recover spectral reflectance with a higher accuracy than other methods considered.

Colour preference is a critical dimension for describing the colour quality of lighting and numerous metrics have been proposed. However, due to the variation amongst psychophysical studies, consensus has not been reached on the best approach to quantify colour preference. In this study, 25 typical colour quality metrics were comprehensively tested based on 39 groups of psychophysical data from 19 published visual studies. The experimental results showed that two combined metrics: the arithmetic mean of the gamut area index (GAI) and colour rendering index (CRI) and the colour quality index (CQI), a combination of the correlated colour temperature (CCT) and memory colour rendering index (MCRI), exhibit the best performance. Qp in the colour quality scale (CQS) and MCRI also performed well in visual experiments of constant CCT but failed when CCT varied, which highlights the dependence of certain metrics on contextual lighting conditions. In addition, it was found that some weighted combinations of an absolute gamut-based metric and a colour fidelity metric exhibited superior performance in colour preference prediction. Consistent with such a result, a novel metric named MCPI (colour preference index based on meta-analysis) was proposed by fitting the large psychophysical dataset, and this achieved a significantly higher weighted average correlation coefficient between metric predictions and subjective preference ratings.

Digitization of cultural heritage protection has received considerable attention in heritage studies and spectral imaging technology has been playing an important role in this research. This article aims to study the technique of selecting optimal filter set to obtain ancient murals spectral image with high spectral and colorimetric accuracy based on the broadband spectral imaging system. The 330 Dunhuang murals mineral pigment color patches and the GretagMacbeth ColorChecker (CC) as well as 27 pieces of optical filters chosen as samples were examined. For each piece of filter, the three‐channel image was captured by the spectral imaging system. Then, 351 groups of six‐channel digital count images were acquired by arbitrary combinations of two among the 27 three‐channel digital count images. The pseudo‐inverse, principal component analysis, and R‐matrix methods were used to reconstruct the spectral reflectance from the six‐channel digital counts for each sample. Finally, this study identified the optimal filter set by evaluating the integrated error (TOTAL ERROR), which was calculated by normalizing the mean spectral root‐mean‐square error (RMS), mean spectral goodness‐of‐fit error (1‐GFC), and mean CIEDE2000 color difference (ΔE00) and by multiplying them together. After the optimal optic filter set was selected, it was applied to the Dunhuang murals spectral imaging and was evaluated. The results showed that the optimal optic filter set could result in promising improvement both in spectral and color accuracy when compared with the production camera. In addition, it can be used for the construction of Dunhuang murals spectral image database. © 2015 Wiley Periodicals, Inc. Col Res Appl, 41, 585–595, 2016