Titlebook: Computer Vision - ACCV‘98; Third Asian Conferen Roland Chin,Ting-Chuen Pong Conference proceedings 1997 Springer-Verlag Berlin Heidelberg 1

SLING

Gudgeon

Optimal edge detection under difficult imaging conditions,tion and we use the dual intensity and line processes introduced by [Geman and Geman, 1984]. The approach seeks to minimize a global energy functional that explicitly incorporates image properties to be minimized into weighted terms of the energy functional. Our specific contribution is modifying th

transdermal

Restoring image quality through structure preserving de-noising,the noise while maintaining good visual quality. This problem has assumed major significance with the increase in image related communication that has accompanied the exponential growth of the internet. Traditionally, image quality is measured in terms of PSNR (Peak Signal to Noise Ratio) which may

滔滔不絕地講

Feature saliency from noise variations in invariants,l objects in complex settings. This requires considerable robustness and reliability, and so low level invariants are used as a robust starting point. In particular, a set of quantities is developed that are both . and . invariant. They are arranged as the components of a description vector, which a

ARIA

燒烤

光滑

Segmenting objects at multiple scales: A robust approach,on system to perform successfully, the segmentation procedure used must be robust in presence of noise and local distortions of shape. Furthermore, it should be based on geometric invariants so that the segmentation will not be affected by arbitrary choices. This paper proposes a new multi-scale seg

boisterous

Multi-grid edge models for magnifying digital images,l step edge patterns. Based on the assumptions that within a small image region, the underlying edge structure and the average pixel intensity of the low and high resolution image samples should be identical, a small window of low resolution pixels are mapped to a high resolution lattice. The method

affinity

Scale and rotation invariant recognition method using higher-order local autocorrelation features o. Linear scalings and rotations are represented as shifts in the log-polar image which is obtained by re-sampling of the input image. HLAC features of log-polar image become robust to the linear scalings and rotations of a target because HLAC features are shift invariant. By combining these features

消耗