The authors develop several versions of the diffusion equation to demosaic color filter arrays of any kind. In particular, they compare isotropic versus anisotropic and linear versus non-linear formulations. Using these algorithms, they investigate the effect of mosaics on the resulting demosaiced images. They perform cross analysis on images, mosaics, and algorithms. They find that random mosaics do not perform the best with their algorithms, but rather pseudo-random mosaics give the best results. The Bayer mosaic also shows equivalent results to good pseudo-random mosaics in terms of peak signal-to-noise ratio but causes visual aliasing artifacts. The linear anisotropic diffusion method performs the best of the diffusion versions, at the level of state-of-the-art algorithms.
A new camera metric is proposed based, in concept, on the vision test devised by Davidson and Hemmendinger. The “D&H Color Rule” is a set of two linear patch arrays containing at most one match for near-normal color observers. The match selected by an observer can provide an indication of how that observer’s color vision relates to others. Comparisons may be made to a group of interest, or more commonly, to a CIE Standard Observer. For this research, two image targets have been created, one physical and one virtual, each with a family of spectra related in much the same way as the colors in the original D&H Rule. These targets can be physically imaged or virtually modeled to predict camera RGB, and then CIELAB with a color profile. The camera can then be judged as to what degree its output matches that of a CIE Standard Observer.
Neugebauer-type models for reflectance factor spectra produced by halftone-based color hardcopy rely on the spectra of the so-called Neugebauer primaries, an invertible function of reflectance, and a weight, or scalar, for each Neugebauer primary. This paper discusses the calculation of the third factor.
Research has, until now, relied heavily, perhaps exclusively, on just three parameter-less sets of equations to compute these scalars. In the previous paper in this series, it was shown how the scalar for each Neugebauer primary could be computed from the function governing the scalar of the primary consisting of all colorants, the so-called final scalar. In this paper, conditions necessary and sufficient for a function to provide the final scalar are recited. Copulas, an entire class of functions from probability theory, also satisfy these conditions, and therefore may be used for final scalar calculation.
Halftone patterns generated by a raster image processor were used to assess potential improvement in final scalar accuracy in two-colorant dot-on-dot printing with a slight amount of misregistration. Using a copula new to this application resulted in a nearly ten-fold improvement in accuracy over the best-performing overlap models in the prior art, indicating significant increases in accuracy are to be expected with actual printed artifacts.
Raw images are more useful than JPEG images for machine vision algorithms and professional photographers because raw images preserve a linear relation between pixel values and the light measured from the scene. A camera is radiometrically calibrated if there is a computational model which can predict how the raw image is mapped to the corresponding rendered image (e.g., JPEGs) and vice versa. Our method makes use of the observation that the rank order of pixel values is mostly preserved post-color correction. We show that this observation is the key for getting a compact and robust radiometric calibration model. Since our method requires fewer variables, it can be solved for using less calibration data. An additional advantage is that we can derive the camera pipeline from a single pair of raw–JPEG images. Experiments demonstrate that our method delivers state-of-the-art results (especially for the most interesting conversion from JPEG to raw).
Remote photoplethysmography (rPPG) enables us to capture the vital signs such as pulse rate, respiratory rate, oxygen saturation, and even heart rate variability (HRV) without any contact devises. Although the papers of rPPG mainly focus on the use of standard RGB camera, it cannot used for the cases in night or under the dim light conditions. Therefore, in this paper, we propose a novel noncontact method for monitoring HRV without visible lighting. The proposed method uses dual-band infrared videos to ensure robustness to fluctuations in illumination. The hemoglobin component is extracted via a simple projection from the dual-band pixel values in logarithmic space. We demonstrate the accurate extraction of pulse wave signals using pulse wave maps. As the results, we indicated the effectiveness of HRV monitoring in the situation under the dim light condition.
The use of multi-spectral imaging has been found to improve the accuracy of deep neural network-based pedestrian detection systems, particularly in challenging night time conditions in which pedestrians are more clearly visible in thermal long-wave infrared bands than in plain RGB. In this article, the authors use the Spectral Edge image fusion method to fuse visible RGB and IR imagery, prior to processing using a neural network-based pedestrian detection system. The use of image fusion permits the use of a standard RGB object detection network without requiring the architectural modifications that are required to handle multi-spectral input. We contrast the performance of networks trained using fused images to those that use plain RGB images and networks that use a multi-spectral input.
In this article, we introduce a method for using a smartphone to estimate the bi-directional reflectance distribution functions of real-world surfaces. Specifically, we use a combination of image-based and visual techniques to determine a surface’s diffuse and specular reflectance parameters in the Ward light reflection model. We test the accuracy of the method by comparing our estimates to instrumental measurements of color and gloss standards. We demonstrate its utility by rendering synthetic images of method-estimated surfaces and comparing the renderings with photographs of the real surfaces. The method provides a lightweight approach to surface reflectance measurement.
The study is the second part of a previous study which explored the effects of color pairs on warmth perception in interiors. The main aim of this study is to investigate the effects of material pairs and their single materials on warmth perception in interiors with the same methodology, since paired materials have not been investigated yet. Each material pair and their two single materials were assessed by 32 different participants, thus 96 different participants assessed three groups of material models (Fabric and Timber material pair, Fabric and Plasterboard material pair, Timber and Plasterboard material pair, and their single materials) under controlled conditions. Results indicated that as single materials Timber and Fabric have the same level of warmth and are warmer than Plasterboard whereas there is no difference between their pairs. Findings revealed that these two natural materials are perceived to be warmer than the artificial one and pairing them on interior walls provides similar level of warmth.