Computing Science, School of

Receive updates for this collection

Automatic White Balancing via Gray Surface Identification

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2007-11
Abstract: 

The key to automatic white balancing of digital imagery is to estimate accurately the color of the overall scene illumination. Many methods for estimating the illumination’s color have been proposed [1-6]. Although not the most accurate, one of the simplest and quite widely used methods is the gray world algorithm [6]. Borrowing on some of the strengths and simplicity of the gray world algorithm, we introduce a modification of it that significantly improves on its performance while adding little to its complexity.

Document type: 
Conference presentation
File(s): 

Skin Colour Imaging that is Insensitive to Lighting Conditions

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2008-06
Abstract: 

In previous human skin models, it has been suggested that the colour of human skin is mostly determined by the concentration of melanin in the epidermal layer combined with the concentration of hemoglobin in the dermal layer. The colour of facial skin changes significantly with changes in the light incident upon it. In this paper we propose a method of normalizing the skin tones of human faces that eliminates the effects of illumination, preserving the skin colour and allowing variations related to melanin concentration only. The method assumes the illumination is reasonably well modelled as blackbody radiation.

Document type: 
Conference presentation
File(s): 

Dichromatic Illumination Estimation via Hough Transforms in 3D

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2008-06
Abstract: 

A new illumination-estimation method is proposed based on the dichromatic reflection model combined with Hough transform processing. Other researchers have shown that using the dichromatic reflection model under the assumption of neutral interface reflection, the color of the illuminating light can be estimated by intersecting the dichromatic planes created by two or more differently coloured regions. Our proposed method employs two Hough transforms in sequence in RGB space. The first Hough Transform creates a dichromatic plane histogram representing the number of pixels belonging to dichromatic planes created by differently coloured scene regions. The second Hough Transform creates an illumination axis histogram representing the total number of pixels satisfying the dichromatic model for each posited illumination axis. This method overcomes limitations of previous approaches that include requirements such as: that the number of distinct surfaces be known in advance, that the image be presegmented into regions of uniform colour, and that the image contain distinct specularities. Many of these methods rely on the assumption that there are sufficiently large, connected regions of a single, highly specular material in the scene. Comparing the performance of the proposed approach with previous non-training methods on a set of real images, the proposed method yields better results while requiring no prior knowledge of the image content.

Document type: 
Conference presentation
File(s): 

An Easily Extensible HMM Word Aligner

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2018-10-24
Abstract: 

In this paper, we present a new word aligner with built-in support for alignment types, as well as comparisons between various models and existing aligner systems. It is an open source software that can be easily extended to use models of users' own design. We expect it to suffice the academics as well as scientists working in the industry to do word alignment, as well as experimenting on their own new models. Here in the present paper, the basic designs and structures will be introduced. Examples and demos of the system are also provided

Document type: 
Article
File(s): 

The Effect of Exposure on MaxRGB Color Constancy

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2010-02
Abstract: 

The performance of the MaxRGB illumination-estimation method for color constancy and automatic white balancing has been reported in the literature as being mediocre at best; however, MaxRGB has usually been tested on images of only 8-bits per channel. The question arises as to whether the method itself is inadequate, or rather whether it has simply been tested on data of inadequate dynamic range. To address this question, a database of sets of exposure-bracketed images was created. The image sets include exposures ranging from very underexposed to slightly overexposed. The color of the scene illumination was determined by taking an extra image of the scene containing 4 Gretag Macbeth mini Colorcheckers placed at an angle to one another. MaxRGB was then run on the images of increasing exposure. The results clearly show that its performance drops dramatically when the 14-bit exposure range of the Nikon D700 camera is exceeded, thereby resulting in clipping of high values. For those images exposed such that no clipping occurs, the median error in MaxRGB’s estimate of the color of the scene illumination is found to be relatively small.

Document type: 
Conference presentation
File(s): 

Cubical Gamut Mapping Colour Constancy

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2010-06
Abstract: 

A new color constancy algorithm called Cubical Gamut Mapping (CGM) is introduced. CGM is computationally very simple, yet performs better than many currently known algorithms in terms of median illumination estimation error. Moreover, it can be tuned to minimize the maximum error. Being able to reduce the maximum error, possibly at the expense of increased median error, is an advantage over many published color constancy algorithms, which may perform quite well in terms of median illumination-estimation error, but have very poor worst-case performance. CGM is based on principles similar to existing gamut mapping algorithms; however, it represents the gamut of image chromaticities as a simple cube characterized by the image’s maximum and minimum rgb chromaticities rather than their more complicated convex hull. It also uses the maximal RGBs as an additional source of information about the illuminant. The estimate of the scene illuminant is obtained by linearly mapping the chromaticity of the maximum RGB, minimum rgb and maximum rgb values. The algorithm is trained off-line on a set of synthetically generated images. Linear programming techniques for optimizing the mapping both in terms of the sum of errors and in terms of the maximum error are used. CGM uses a very simple image pre-processing stage that does not require image segmentation. For each pixel in the image, the pixels in the Nby- N surrounding block are averaged. The pixels for which at least one of the neighbouring pixels in the N-by-N surrounding block differs from the average by more than a given threshold are removed. This pre-processing not only improves CGM, but also improves the performance of other published algorithms such as max RGB and Grey World.

Document type: 
Conference presentation
File(s): 

Color Calibration via Natural Food Colors

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2010-10
Abstract: 

Color image calibration is usually done with the aid of a color chart such as the Macbeth ColorChecker containing a set of carefully produced color patches. However, in many consumer applications such as Internet shopping, for which the correct reproduction of color can be very important, most users will not have a color chart readily available, and probably are not interested in purchasing one in any case. We propose using the colors of the fleshy interior parts of oranges, lemons and limes, along with cooked egg white as a means of creating a simple color ‘chart’. A sample of oranges, lemons and limes from North America and Australia has shown their color to be quite consistent, and therefore potentially suitable as a set of reference colors for color image calibration. Figure 1 shows one of the images used in measuring the colors of the fruits and vegetables. In the case of Internet sales, a seller photographing color-sensitive merchandise, such as clothing, could simply include one or two of these foods in each picture. This would provide an immediate point of reference for the purchaser as to whether or not the image colors are correct. Clearly, if the food colors do not look right, neither will the merchandise when it is delivered.

Document type: 
Conference presentation
File(s): 

The Rehabilitation of MaxRGB

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2010-11
Abstract: 

The poor performance of the MaxRGB illuminationestimation method is often used in the literature as a foil when promoting some new illumination-estimation method. However, the results presented here show that in fact MaxRGB works surprisingly well when tested on a new dataset of 105 high dynamic range images, and also better than previously reported when some simple pre-processing is applied to the images of the standard 321 image set [1]. The HDR images in the dataset for color constancy research were constructed in the standard way from multiple exposures of the same scene. The color of the scene illumination was determined by photographing an extra HDR image of the scene with 4 Gretag Macbeth mini Colorcheckers at 45 degrees relative to one another placed in it. With preprocessing, MaxRGB’s performance is statistically equivalent to that of Color by Correlation [2] and statistically superior to that of the Greyedge [3] algorithm on the 321 set (null hypothesis rejected at the 5% significance level). It also performs as well as Greyedge on the HDR set. These results demonstrate that MaxRGB is far more effective than it has been reputed to be so long as it is applied to image data that encodes the full dynamic range of the original scene.

Document type: 
Conference presentation
File(s): 

Gaussian-Metamer-Based Prediction of Colour Stimulus Change under Illuminant Change

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2011-06
Abstract: 

Predicting how the LMS cone response to light reflected from a surface changes with changing lighting conditions is a long-standing and important problem. It arises in white balancing digital imagery, and when re-rendering printed material for viewing under a second illuminant (e.g., changing from D65 to F11). Von Kries scaling is perhaps the most common approach to predicting what LMS cone response will arise under a second illuminant given the LMS under a first illuminant. We approach this prediction problem, instead, from the perspective of Logvinenko’s new colour atlas, and obtain better results than with von Kries scaling.

Document type: 
Conference presentation
File(s): 

Intersecting Color Manifolds

Peer reviewed: 
Yes, item is peer reviewed.
Date created: 
2011-11
Abstract: 

Logvinenko’s color atlas theory provides a structure in which a complete set of color-equivalent material and illumination pairs can be generated to match any given input RGB color. In chromaticity space, the set of such pairs forms a 2-dimensional manifold embedded in a 4-dimensional space. For singleilluminant scenes, the illumination for different input RGB values must be contained in all the corresponding manifolds. The proposed illumination-estimation method estimates the scene illumination based on calculating the intersection of the illuminant components of the respective manifolds through a Hough-like voting process. Overall, the performance on the two datasets for which camera sensitivity functions are available is comparable to existing methods. The advantage of the formulating the illumination-estimation in terms of manifold intersection is that it expresses the constraints provided by each available RGB measurement within a sound theoretical foundation.

Document type: 
Conference presentation
File(s):