A new algorithm of subpixel corner detection;

The exact corner detection is vital to image processing and measurement.

This paper expounds corner detection method to get the position information of every braided unit's climaxes.

Based on the analysis of the characteristics of the spline wavelet representing images in the scale-space,spline wavelet technique has been researched on the demand of image enhancement in corner extraction of image segmentation.

A new wavelet transform approach was presented for local feature extraction (mainly contour extraction and corner extraction),and the feasibility and effectiveness of the proposed approach were demonstrated experimentally.

Second, to meet the requirement of high accuracy in some image registration fields, a algorithm based on corner extraction and correlation matching is used.

Its two distinctive features are proposing a new elementary line segment extraction algorithm by utilizing template and taking into account both traits of the lines and the objects shape during performing line segment grouping.

Divides the edge into several sub-edges in one direction and detect lines from sub-edges according to the line extraction criterions.

For feature extraction from multi-source images,a new line extraction algorithm is developed by fusing the edge information of multi-source images.

It calculated the coordinate of the aircraft in the 3D photogrammetric coordinate system according to two 2D digital photos by using the geometrical symmetry of aircraft,after edge detection and line extraction.

There are many approaches to extracting straight lines, but they all have their particular drawbacks to limit their use.