论文信息 - Pattern analysis on shift, rotation, and scaling

Pattern analysis on shift, rotation, and scaling

Numerous methods for recognising patterns invariant to their rotation, shift and scaling within an image have been proposed. These methods usually suffer from a lack of information as to the geometric feature estimation of tested patterns. In pattern analysis, this information is quite important. The Letter presents an improved algorithm applied to pattern classification and geometric feature estimation with good results.

Yung-Sheng Chen | W.-S. Shao

[1] D. Casasent,et al. Position, rotation, and scale invariant optical correlation. , 1976, Applied optics.

[2] Z. Q. Wang,et al. Amplitude-compensated matched filtering. , 1988, Applied optics.

[3] Zhi-Qiang Liu,et al. On the minimum number of templates required for shift, rotation and size invariant pattern recognition , 1988, Pattern Recognit..