We describe an analysis method of the omnidirectional color signals in natural scenes. A multiband imaging system with six spectral channels is used for capturing high resolution images in the omnidirectional observations at three locations on campus. The spectral distributions of color signals are recovered using the Wiener estimator from the captured six-band images. The spectral compositions of omnidirectional color signals are investigated based on the PCA of each set of color signals acquired at three locations in different seasons and different times of the day. Three principal components are extracted from three sets of omnidirectional images observed in three different locations. The respective three principal component curves are invariant under seasonal and temporal changes. Moreover, we determine the unified principal components of color signals across all locations. High data compression of omnidirectional images can be achieved. The reliability of the proposed analysis method is confirmed using various experimental data.
We propose a method for automatically classifying multiple objects in a natural scene into metal or dielectric. We utilize
polarization property in order to classify the objects into metal and dielectric, and surface-spectral reflectance in order to
segment the scene image into different object surface regions. An imaging system is developed using a liquid crystal
tunable filter for capturing both polarization and spectral images simultaneously. Our classification algorithm consists of
three stages; (1) highlight detection based on luminance threshold, (2) material classification based on the spatial
distribution of the degree of polarization at the highlight area, and (3) image segmentation based on illuminant-invariant
representation of the spectral reflectance. The feasibility of the proposed method is examined in detail in experiments
using real-world objects.
This paper proposes a method for analyzing the color characteristics of woodblock prints having oil-based ink and
rendering realistic images based on camera data. The analysis results of woodblock prints show some characteristic
features in comparison with oil paintings: 1) A woodblock print can be divided into several cluster areas, each with
similar surface spectral reflectance; and 2) strong specular reflection from the influence of overlapping paints arises only
in specific cluster areas. By considering these properties, we develop an effective rendering algorithm by modifying our
previous algorithm for oil paintings. A set of surface spectral reflectances of a woodblock print is represented by using
only a small number of average surface spectral reflectances and the registered scaling coefficients, whereas the previous
algorithm for oil paintings required surface spectral reflectances of high dimension at all pixels. In the rendering process,
in order to reproduce the strong specular reflection in specific cluster areas, we use two sets of parameters in the
Torrance-Sparrow model for cluster areas with or without strong specular reflection. An experiment on a woodblock
printing with oil-based ink was performed to demonstrate the feasibility of the proposed method.
A lighting system is proposed to render objects under a variety of colored illumination. The proposed system is
constructed with a LED unit, white diffusion filters, dimmers, and a personal computer as a controller. The LED unit is
composed of four kinds of color LED lamps which are 12 red (R), 14 green (G), 12 blue (B) and 10 white (W) colors.
The LED lamps have a linear input-output relationship and a larger color gamut than Adobe RGB. Since the lighting
system has an independent white light source, white illumination can be produced using the white light source and a
mixture of RGB primary sources. Therefore, to determine illumination color we have to solve a mapping problem from
3D color space to 4D space of RGBW digital values. This paper proposes an effective algorithm for determining the
digital control signals of the RGBW lights, so that colored light is generated with arbitrary (x, y) chromaticity and
luminance value Y. The performance of proposed method is examined in an experiment, where the accuracy of the
colored light is evaluated with regard to the CIE color difference.
We propose a spectral imaging method for material classification and inspection of raw printed circuit boards (PCBs). The method is composed of two steps (1) estimation the PCB surface-spectral reflectances and (2) unsupervised classification of the reflectance data to make the inspection of PCB easy and efficient. First, we develop a spectral imaging system that captures high dynamic range images of a raw PCB with spatially and spectrally high resolutions in the region of visible wavelength. The surface-spectral reflectance is then estimated at every pixel point from multiple spectral images, based on the reflection characteristics of different materials. Second, the surface-spectral reflectance data are classified into several groups, according to the number of PCB materials. We develop an unsupervised classification algorithm incorporating both spectral information and spatial information, based on the Nystrom approximation of the normalized cut method. The initial seeds for the Nystrom procedure are effectively chosen using a guidance module based on the K-means algorithm. Low-dimensional spectral features are efficiently extracted from the original high-dimensional spectral reflectance data. The feasibility of the proposed method is examined in experiments using real PCBs in detail.
This paper proposes a method for real-time color measurement using active illuminant. A synchronous measurement
system is constructed by combining a high-speed active spectral light source and a high-speed monochrome camera. The
light source is a programmable spectral source which is capable of emitting arbitrary spectrum in high speed. This
system is the essential advantage of capturing spectral images without using filters in high frame rates. The new method
of real-time colorimetry is different from the traditional method based on the colorimeter or the spectrometers. We
project the color-matching functions onto an object surface as spectral illuminants. Then we can obtain the CIE-XYZ
tristimulus values directly from the camera outputs at every point on the surface. We describe the principle of our
colorimetric technique based on projection of the color-matching functions and the procedure for realizing a real-time
measurement system of a moving object. In an experiment, we examine the performance of real-time color measurement
for a static object and a moving object.
This paper describes the investigation and analysis of color terms in modern Japanese. Japanese people use a large
vocabulary of color terms in Japanese unconsciously in their daily life. The authors have studied the basic color terms.
The color vocabulary was investigated for modern Japanese over 6 years, and in all 2,100 subjects participated in the
vocabulary test. This paper shows the investigation process and analyzes the collected color vocabularies from various
points of view. The vocabulary test is based on a questionnaire format without showing any color samples, where each
subject was requested to answer the question items for color names in two levels of importance. Therefore we collected
their recall color names without a priori clue such as color samples. The frequency of occurrence of the responded color
names is statistically evaluated, and then the importance of color names is analyzed from various points of view.
An eyegaze interface is one of the key technologies that serves as an input device in a ubiquitous-computing society. Recently, video-based techniques that do not require specific instruments have been studied. With these approaches, development of an accurate iris-extraction algorithm is very important to realize practical eyegaze tracking. For accurate iris extraction, it is necessary to achieve robustness, high speed, and high accuracy. Conventional iris-extraction algorithms experience difficulties in meeting all these requirements simultaneously. This study proposes an iris-extraction algorithm based on the parametric template matching method to satisfy all these requirements at the same time. The parametric template matching method achieves robustness by interpolating among some templates, and the method attains high accuracy by a subpixel matching technique. High-speed matching can be realized by coarse-to-fine matching. To verify the effectiveness of the proposed algorithm, we performed a basic experiment for eyegaze tracking. We show in this experiment that the processing time is approximately 1/900 of that of our previous method and that accuracy is acceptable with the new method. Then, we apply the proposed algorithm to an eyegaze keyboard, along with an imaging system for improving image quality, and we verify the effectiveness of this approach.
A method is proposed for estimating the spectral reflectance function of an object surface by using a six-color scanner.
The scanner is regarded as a six-band spectral imaging system, since it captures six color channels in total from two
separate scans using two difference lamps. First, we describe the basic characteristics of the imaging systems for a HP
color scanner and a multiband camera used for comparison. Second, we describe a computational method for recovering
surface-spectral reflectances from the noisy sensor outputs. A LMMSE estimator is presented as an optimal estimator.
We discuss the reflectance estimation for non-flat surfaces with shading effect. A solution method is presented for the
reliable reflectance estimation. Finally, the performance of the proposed method is examined in detail on experiments
using the Macbeth Color Checker and non-flat objects.
This paper proposes a computer vision system for improving the image quality around a steady gaze point on a display
device. We assume that one observes a localized small area of the displayed image, rather than the whole image, because
of a limited visual angle. The computer vision system consists of two subsystems which are an eyegaze detection
subsystem and an image quality improvement subsystem. The eyegaze detection subsystem tracks a human gaze point on
the display. A tracking algorithm is developed for capturing a human face from a single monocular camera without using
any special devices. The image quality improvement subsystem performs a localized Retinex algorithm. Although the
conventional algorithms contain a large number of complex computations, the localized algorithm is devised for
performing the Retinex computation in high speed for only a localized part within the whole image. The combined
system is developed so that the image quality is improved in real time within just local region around the detected gaze
point. We make an experimental system consisting of an off-the-shelf digital video camera and a personal computer. The
whole performance of the computer vision system is examined experimentally on subjective assessment and processing