MUCKE aims to mine a large volume of images, to structure them conceptually and to use this conceptual structuring in order to improve large-scale image retrieval. The last decade witnessed important progress concerning low-level image representations. However, there are a number problems which need to be solved in order to unleash the full potential of image mining in applications. The central problem with low-level representations is the mismatch between them and the human interpretation of image content. This problem can be instantiated, for instance, by the incapability of existing descriptors to capture spatial relationships between the concepts represented or by their incapability to convey an explanation of why two images are similar in a content-based image retrieval framework. We start by assessing existing local descriptors for image classification and by proposing to use co-occurrence matrices to better capture spatial relationships in images. The main focus in MUCKE is on cleaning large scale Web image corpora and on proposing image representations which are closer to the human interpretation of images. Consequently, we introduce methods which tackle these two problems and compare results to state of the art methods. Note: some aspects of this deliverable are withheld at this time as they are pending review. Please contact the authors for a preview.

http://ieeexplore.ieee.org/iel2/4524/12820/00592460.pdf

Image Processing

Color image segmentation: advances and prospects

One of the challenging issues for Orthogonal Frequency Division Multiplexing (OFDM) system is its high Peak-to-Average Power Ratio (PAPR). In this paper, we review and analysis different OFDM PAPR reduction techniques, based on computational complexity, bandwidth expansion, spectral spillage and performance. We also discuss some methods of PAPR reduction for multiuser OFDM broadband communication systems.

https://engold.ui.ac.ir/~sabahi/Advanced%20digital%20communication/An%20Overview%20Peak-to-Average%20Power%20Ratio%20Reduction%20Reduction%20Techniques%20for%20OFDM%20Signals.pdf

An Overview: Peak-to-Average Power Ratio Reduction Techniques for OFDM Signals

Computer and Robot Vision Vol. 1, by R.M. Haralick and Linda G. Shapiro, Addison-Wesley, 1992, ISBN 0-201-10887-1.

Computer and Robot Vision

This paper presents a new reflectance model for rendering computer synthesized images. The model accounts for the relative brightness of different materials and light sources in the same scene. It describes the directional distribution of the reflected light and a color shift that occurs as the reflectance changes with incidence angle. The paper presents a method for obtaining the spectral energy distribution of the light reflected from an object made of a specific real material and discusses a procedure for accurately reproducing the color associated with the spectral energy distribution. The model is applied to the simulation of a metal and a plastic.

A reflectance model for computer graphics

Projectors have become common display devices, not only for office and school presentations, but also for home theater
entertainment. Although a completely dark room is the ideal venue for watching a projected image, in most situations
(including classrooms and conference rooms) the viewing conditions are not completely dark, and ambient light falling
on the screen produces a background light level with the image projected on top. As the background light increases, it
becomes more difficult to see the projected image, which becomes dull and may appear washed out. What is really
happening is that the ambient light reduces the contrast of the image. While the amount of light contributing to the image
remains the same, more light has been projected onto the screen by other light sources. This effect can be reduced by
employing the white-peaking function of a digital light-processing (DLP) projector, which adjusts the white segment of
the color wheel, resulting in more natural and vivid images. Although the chromaticity coordinates for an image
projected with and without white peaking are the same, when white is added to the projected image, the perceived hue
changes. This phenomenon is known as the Abney effect. This paper presents a model of this hue-shift phenomenon and
proposes a hue-correction method. For evaluation purposes, an observer-preference test is conducted on several test
images with and without hue shifts, and z-scores are utilized to compare the results.

Hue-shift model for DLP projector with the white peaking function

Recent research efforts have focused on combining high dynamic range (HDR) imaging with super-resolution (SR) reconstruction to enhance both the intensity range and resolution of images beyond the apparent limits of the sensors that capture them. The processes developed to date start with a set of multipleexposure input images with low dynamic range (LDR) and low resolution (LR), and require several procedural steps: conversion from LDR to HDR, SR reconstruction, and tone mapping. Input images captured with irregular exposure steps have an impact on the quality of the output images from this process. In this paper, we present a simplified framework to replace the separate procedures of previous methods that is also robust to different sets of input images. The proposed method first calculates weight maps to determine the best visible parts of the input images. The weight maps are then applied directly to SR reconstruction, and the best visible parts for the dark and highlighted areas of each input image are preserved without LDR-to-HDR conversion, resulting in high dynamic range. A new luminance control factor (LCF) is used during SR reconstruction to adjust the luminance of input images captured during irregular exposure steps and ensure acceptable luminance of the resulting output images. Experimental results show that the proposed method produces SR images of HDR quality with luminance compensation. Introduction Many efforts have been made to enhance the image quality of digital still cameras by improving the physical performance of their image sensors. Even so, digital cameras still suffer from limited dynamic range and resolution that is less than that encountered in the real world. High dynamic range (HDR) imaging algorithms have been developed to overcome the problem of underexposed or overexposed images caused by the narrow dynamic range of cameras. This involves assembling multiple-exposure low dynamic range (LDR) images from a normal camera to obtain a full dynamic range image [1]. However, HDR imaging techniques require that the camera response curve (CRC) for each camera is capable of recovering the intensity of an actual scene [2]. Additionally, many common displays have a limited dynamic range and cannot display HDR images directly; such displays require a tone-mapping process to compresses the dynamic range of the image to fit their dynamic ranges [2]. Many tone-mapping algorithms have been proposed, but each is tailored to a specific purpose and thus has its own advantages and disadvantages. Thus, the quality of HDR images is not always preserved. The limited resolution of normal digital cameras has been overcome with SR reconstruction, which increases the spatial resolution by exploiting the correlation of several sequential input images obtained by the camera under identical conditions. Because SR reconstruction requires the use of multiple images with the same exposure time, combining it with HDR imaging is difficult because the later requires multiple-exposure images. Even so, recent studies have addressed the challenge of combining HDR images and SR reconstruction to obtain high-quality, highresolution images with high dynamic range. Gunturk sought to obtain HDR-SR images, proposing a new imaging model during SR reconstruction that included dynamic range and spatial domain effects [3]. Figure 1 and the following equation describe that imaging model: ( ) k k k k k f t q v = + + z H w (1) where zk is a low-resolution observation for the k LDR-LR image with Y channel from YCbCr values, f(·) is the nonlinear camera response function, tk is the exposure time, q is a high-resolution input signal, and Hk is the linear mapping that incorporates motion, the point spread function, and down-sampling. Although the resulting images contain HDR information, presenting them on normal displays requires tone mapping, which may not give satisfactory results depending on the tone mapping method used. Schubert also suggested a framework for combining HDR and SR [4]. He used a similar imaging model that included photometric camera calibration data obtained using Debevec’s method and tone mapping [2]. Both methods require estimating the CRC function to obtain the dynamic range of the actual scene; however, tone mapping is also required and produces results that are not always satisfactory. Indeed, estimation of the real-world CRC function is not a simple task. Furthermore, each camera requires its own CRC function, and tone-mapping algorithms have a very significant influence on the quality of the resulting images. Here, we propose a new framework for obtaining high-quality images that appear to have high dynamic range and super resolution. Because it blends multiple-exposure images, the framework is robust in the face of input images from irregular exposure steps. The proposed method first calculates weight maps, Figure 1. Imaging model by Gunturk. which it then applies directly to SR reconstruction with pyramid merging [5, 6]. These weight maps are obtained by retaining only 19th Color and Imaging Conference Final Program and Proceedings 195 the best visible parts in each multiple-exposure input image to preserve details of dark and highlighted regions during SR reconstruction and produce SR output images that have HDR quality without LDR-to-HDR conversion [5]. However, the use of weight maps alone does not result in suitable luminance in output images because of the set of input images from irregular exposure steps. Thus, a new luminance control factor (LCF) is proposed for application during SR reconstruction to correct this. The resulting images have suitable resolution in both light and dark sections. SR reconstruction based on weight maps The overall objective of this work was to obtain an SR image that has fine details in light and dark sections. Figure 2 is a flowchart for the proposed method assuming four input images with proper exposure steps. The specific procedure has two parts: color reproduction and SR reconstruction from multiple-exposure images. During color reproduction, exposure fusion is performed on a series of LDR-LR input images to produce a single HDR-LR image [5]. Then, only CbCr values are extracted and resized for use in the last conversion from YCbCr to digital RGB values. Next, SR reconstruction takes place assuming perfect image registration. During this process, weight maps are first calculated to determine the necessary areas in each input image that contain the “best” intensities. These weight maps are then applied to SR reconstruction with pyramid merging to prevent intensity distortion from unwanted iterations. Simultaneously, LCFs (ηk) are applied to each input image to adjust the luminance so that the overall luminance of the output image is suitable. Weight map construction The exposure fusion algorithm generates high-quality images such as those that result from HDR imaging [5]. That is, the resulting images show fine details in dark and lighted areas. This algorithm selects the “best” parts of images in a multi-exposure image sequence. These “best” parts are defined as a weighted map Figure 2. Flow of the proposed method. based on a combination of quality measures, such as contrast, saturation, and good exposure [5], and this map is used to blend the input images. The contrast measure C is calculated as the absolute value of the Laplacian-filtered image for each gray-scaled image. The saturation measure S is obtained as the standard deviation within the R, G, and B channels at each pixel. The measure of good exposure E indicates how well a pixel is exposed, and it is used as a weight of the intensity, based on how close it is to 0.5 using a Gaussian curve defined as e. This is applied to each channel. The final weight maps for each k (k = 1, .., N) image are defined as follows: , , , , ij k ij k ij k ij k W C S E = × × (2) where i and j are the location of pixels, and k is the index of the input sequence images. Figure 3. A set of test input images. Figure 4. Weight maps for each input image. 196 ©2011 Society for Imaging Science and Technology Figure 3 shows a test set of multiple-exposure image sequences with RGB channels with Tiff format. Because the resulting image is affected by the set of input images, seven images were prepared from the auto-exposed image using a Canon 5D Mark 2 camera with fixed aperture value of 5.0 for use a as reference image. Figure 4 shows the results of weight maps for each input image. Dark and highlighted areas have low values on the weight maps; high values indicate the “best” areas of the input images. Proposed SR reconstruction We first assume the imaging model to be the following: k k k z q v = + H (3) where zk, Hk, q, vk, and k are the same from equation (1). If the image registration is perfect, then Hk is composed of spatial warping, blurring, and down-sampling. In the deterministic approach of SR reconstruction, the inverse imaging model can be solved by choosing q to minimize the following cost function:

Reconstruction of Super Resolution High Dynamic Range Image from Multiple-Exposure Images.

A spatially variant B-spline interpolation function is proposed to reduce the blurring appearances caused by the conventional cubic B-spline interpolation function and enhance the local edges that are little considered in the previous methods. We also consider the local characteristics-the difference level between smooth regions and edges and then interpolate the image using the proposed function that is represented by a parametric expression. The proposed function provides the smoothing effects at the smooth regions and adaptively edge-enhanced simulation results compared to another interpolation methods.

Spatially variant B-spline function for image interpolation

This paper proposes a driving current control method for a back light unit (BLU), consisting of red, green, and blue (RGB) light-emitting diodes (LEDs), whereby an RGB optical sensor is used to check the output color stimulus variation to enable a time-stable color stimulus for light emission by the RGB LED BLU. First, to obtain the present color stimulus information of the RGB LED BLU, an RGB to XYZ transform matrix is derived to enable CIEXYZ values to be calculated for the RGB LED BLU from the output values of an RGB optical sensor. The elements of the RGB to XYZ transform matrix are polynomial coefficients resulting from a polynomial regression. Next, to obtain the proper duty control values for the current supplied to the RGB LEDs, an XYZ to Duty transform matrix is derived to calculate the duty control values for the RGB LEDs from the target CIEXYZ values. The data used to derive the XYZ to Duty transform matrix are the CIEXYZ values for the RGB LED BLU estimated from the output values of the RGB optical sensor and corresponding duty control values applied to the RGB LEDs for the present, first preceding, and second preceding sequential check points. With every fixed-interval check of the color stimulus of the RGB LED BLU, the XYZ to Duty transform matrix changes adaptively according to the present lighting condition of the RGB LED BLU, thereby allowing the RGB LED BLU to emit the target color stimulus in a time-stable format, regardless of changes in the lighting condition of the RGB LEDs. Introduction The interest for using light-emitting diodes (LEDs) for display and illumination applications has been growing steadily over the past few years. The potential for long life-time, reduced energy consumption, robustness, and with no Hg are positive attributes of this rapidly evolving technology that have generated so much interest for its use in a variety of applications [1]. The use of LEDs for the backlighting in liquid crystal display (LCD) applications offers more interesting advantages. The performance of color reproduction for the red, green, and blue (RGB) LED backlight is better than that of coldcathode fluorescent lamps (CCFLs) backlight. The color reproduction capability is expressed as a color gamut which is generally depicted by a triangle of primary colors in displays. The value of a color gamut is simply expressed as a percentage between the area of triangle of a display and the area of triangle of a reference color space, such as the National Television Systems Committee (NTSC) standard. The color gamut of the RGB LED backlight can cover nearly 100% of the NTSC [2]. Essentially, since LCDs are hold-type displays while cathode ray tubes (CRTs) are impulse-type displays, motion blur is a more serious problem in LCDs than it is in CRTs. To reduce motion blur in LCDs, the LED backlighting can be helpful used for high frequency driving, in the technique of backlight blinking [3]. Backlight modulation not only works for improving motion blur, but also contributes to a higher contrast ratio and lower power consumption. A whole screen is divided into some areas and LEDs are assembled on each one with independent brightness control operation. By actively controlling the area-brightness of the LEDs corresponding to the display input signals, the contrast ratio of LCDs can be improved [4]. Although LEDs offer various advantages for backlighting, the use of LEDs as a backlighting source has several limitations, one of which is lighting stability. Due to inherent characteristics, the lighting produced by LEDs depends on the temperature and lighting time. The intensity of the output light of LEDs increases with low temperature while it decreases with high temperature. Therefore, under the condition that the temperature of LEDs can not be a constant, it is difficult to make the intensity of the LED backlighting stable. Also, according to the temperature of LEDs, the wavelength of the peak in the spectral power distribution of the LED light shift. Because the directions and amounts of shifted wavelength of the red, green, and blue LED are different from each other, the chromaticity of the RGB LED BLU can be changed by the temperature variation. Moreover, LEDs depend on their lighting time. As the lighting time for LEDs increases, the intensity of the output light of LEDs decreases [5], [6]. Accordingly, for a time-stable RGB LED BLU output, this paper proposes a driving current control method by generating proper duty control values under the present lighting condition of an RGB LED BLU. To check the present output color stimulus of the RGB LED BLU, the forward characterization of the RGB optical sensor is performed. According to the checked present lighting condition of the RGB LED BLU, the proper duty control values are generated using a real-time-update XYZ to Duty transform matrix. Experimental results demonstrate that the proposed duty control method enables a time-stable RGB LED BLU emission. Time-Stability for RGB LED BLU To observe the variation of the output color stimulus for the RGB LED back light unit (BLU) with time, the duty control values for the RGB LEDs are fixed to (2800, 2000, 3200) where the maximum duty control value is 4095 in 12-bit format. Then, the RGB LED BLU was turned on and the output CIEXYZ color stimulus values and the spectral power distribution for the RGB LED BLU lighting were measured with a constant time interval of 10 minutes using a Minolta CS1000 spectroradiometer. As shown in Table 1, the measured CIEXYZ values for the RGB LED BLU are changing continuously with time. After roughly 90 minutes, the CIEXYZ values become stable. This means the temperature of the RGB LED BLU does not change after 90 minutes, consequently, the variation of the color stimulus of the RGB LED BLU caused by 336 ©2008 Society for Imaging Science and Technology changes of temperature does not occur after that time. If the temperature condition of the RGB LED BLU is changed by an internal or external factor, the constant output color stimulus of the RGB LED BLU cannot be guaranteed however. Figure 1 shows the different spectral power distributions of RGB LED BLU with time. Table 1. Change of CIEXYZ values of RGB LED BLU with time Time [minutes] X Y Z 0 6464 6469 11710 10 6308 6334 11610 20 6202 6242 11550 30 6125 6175 11510 : : : : 90 5818 5901 11410 100 5818 5901 11410 110 5818 5901 11410 120 5817 590

Time-Stable RGB LED Backlighting Control Using Time-Varying Transform Matrix.

This paper proposed robust color constancy method for changing illuminant by using local chromaticity distribution and analysis of illuminant influence for each hue angle. First, changing in chromaticity distribution direction for each color with respect to various illuminant is analyzed using principal component analysis. Next, change in standard deviation of chromaticity distribution with respect to various illuminant is also analyzed. Experimental result shows that the proposed method enhances illumination estimation results in terms of angular error between real illuminant and estimated illuminant.

Yeong-Ho Ha

Papers

Hue-shift model for DLP projector with the white peaking function

Reconstruction of Super Resolution High Dynamic Range Image from Multiple-Exposure Images.

Spatially variant B-spline function for image interpolation

Time-Stable RGB LED Backlighting Control Using Time-Varying Transform Matrix.

Color constancy method based on local chromaticity distribution and illuminant influence for hue angle