Search

In this paper, a new illumination invariant color model HSy with large discriminating power is proposed for object recognition in autonomous robot soccer. Object recognition for soccer robot is used as a benchmark problem for the proposed color model. Classical illumination invariant color components are analyzed. New illumination invariant color components are defined based on the dichromatic reflection model. A new color model HSy is constructed by selecting the less correlated components invariant to illumination. The proposed color model is invariant to illumination intensity in the context of the autonomous robot soccer and it yields maximal color discrimination for object recognition. Experiments have been carried out on the soccer robot M-TR for the ball recognition. Experimental results show that the proposed model achieves high object recognition accuracy. [ABSTRACT FROM AUTHOR]

The authors explored a method to control dynamic RGB light source in accordance with human visual environment and possibilities of physical interaction with light source of full color capabilities. We developed a color model composed of hue, grayscale, brightness components and an algorithm to transform RGB values into the color space of the model. We implemented the color model using full color light emitting diode (LED) in a light source device designed for people to interact with color physically. We set up an experimental environment for human color perception to prove visual and physical interaction effect of dynamic light source and object with color composition governed by the color model and the device. [ABSTRACT FROM AUTHOR]

In this paper, we propose an unsupervised color image segmentation scheme using homotopy continuation method and Compound Markov Random Field (CMRF) model. The proposed scheme is recursive in nature where model parameter estimation and the image label estimation are alternated. Ohta (I1, I2, I3) model is used as the color model for image segmentation and we propose a compound MRF model taking care of intra-color and inter-color plane interactions. The CMRF model parameters are estimated using Maximum Conditional Pseudo Likelihood (MCPL) criterion and the MCPL estimates are obtained using homotopy continuation method. The image label estimation is formulated using Maximum a Posteriori criterion and the MAP estimates are obtained using hybrid algorithm. In the context of misclassification error, the proposed unsupervised scheme with CMRF model exhibited improved segmentation accuracy as compared to MRF model and Kato΄s method. [ABSTRACT FROM AUTHOR]

We propose a measure of patient alignment in a video by combining different image representations : grey level, edges, and a set of feature points. When patient head is correctly positionned, a reference image with its ellipse is stored as a template of correct alignment. Edges detection results in a second template of the correct head location. Corners inside the ellipse are detected and tracked: a set of N feature points composes a third template. Template matching computes a measure of similarity between a representation of the reference image and a window sliding around the reference location. Similarity with these three models are combined by the product rule. Location of window the most similar to the templates gives the translation T of the reference model in the image plane. This measure of patient misalignment could avoid X-ray verification of patient alignment, reducing patient dose and duration of treatment sessions. [ABSTRACT FROM AUTHOR]

The performance in object detection under different values of illumination intensity is studied using two major color systems; that is, RGB (Red, Green, and Blue) and HSV (Hue, Saturation, and Value). The comparisons between these two systems are given. The technique called background subtraction static method is adopted in the detection process. Experimentally, it is found that HSV color system is very suitable in object detection applications even under low illumination intensity condition. Finally, it is shown that not all three parameters in HSV color system are useful in the detection. Saturation (S) and Value (V) are the two key parameters to be used in the detection. [ABSTRACT FROM AUTHOR]

Visual perception modeling for intelligent avatars can play subsidiary roles in controlling the behavior and animation of avatars in Intelligent Virtual Environments (IVE). We present a visual perception recognition model by integrating the color model, 3D object matching, sensors perception and path planning, therefore, it is quite useful for the avatar to make appropriate decisions in real-time in an IVE. In our approach, the 3D objects in the IVE are recognized and the moving directions of the avatar are obtained by using the avatar΄s perception viewpoint and sensor. The avatar΄s path planning employs our motion path synthesis proposed previously, called two-stage process synthesis. Experimental results show the efficiency and high prospect of our approach. [ABSTRACT FROM AUTHOR]