Narrow Results

In this paper, robust human detection is investigated by fusing the stereo and infrared thermal images for effective interaction between humans and socially interactive robots. A scale-adaptive filter is first designed for the stereo vision system to detect human candidates. To eliminate the difficulty of the vision system in distinguishing human beings from human-like objects, the infrared thermal image is used to solve the ambiguity and reduce the illumination effect. Experimental results show that the fusion of these two types of images gives an improved vision system for robust human detection and identification, which is the most important and essential component of human robot interaction.

The problem of estimating quantization error in 2D images is an inherent problem in computer vision. The outcome of this problem is directly related to the error in reconstructed 3D position coordinates of an object. Thus estimation of quantization error has its own importance in stereo vision. Although the quantization error cannot be controlled fully, still statistical error analysis helps us to measure the performance of stereo systems that relies on the imaging parameters. Generally, it is assumed that the quantization error in 2D images is distributed uniformly that need not to be true from a practical aspect. In this paper, we have incorporated noise distributions (Triangular and Trapezoidal) for the stochastic error analysis of the quantization error in stereo imaging systems. For the validation of the theoretical analysis, the detailed simulation study is carried out by considering different cases.