Localization of Objects in Noisy Scenes for Robotics Applications Using Wigner Distribution
暂无分享,去创建一个
This paper presents a new method for localization of objects in a noisy scene using Wigner Distribution (WD). We have superimposed noise as high 100% random noise as well as upto -10 dB Signal to Noise Ratio (SNR). With such extreme amount of noise, it is difficult to locate the object with our eyes. However, with the proposed method we can locate the object. The Wigner distribution has been extensively used for speech processing. However, its use for image processing is relatively new. The proposed method can be used for robotics application, where a robot has to locate the objects in highly noisy scenario. The method can also be utilized for finding out the path for robot movement in a cluttered environment. The results indicate that the proposed method deals efficiently with high noise and segments out the objects from the scene.
[1] J. Mayer,et al. On the Quantum Correction for Thermodynamic Equilibrium , 1947 .
[2] E. Wigner. On the quantum correction for thermodynamic equilibrium , 1932 .
[3] Robert J. Marks,et al. The use of cone-shaped kernels for generalized time-frequency representations of nonstationary signals , 1990, IEEE Trans. Acoust. Speech Signal Process..
[4] Robert M. Haralick,et al. Wigner Distribution for 2D Motion Estimation from Noisy Images , 1993, J. Vis. Commun. Image Represent..