According to the characteristics of well-logging graphs, we contribute a PCTR (preprocessing, compressing, tracing, and rectifying) approach to digitizing skeleton type of curves. The preprocessing algorithm is based on the biggest variance ratio. We employed line adjacency graph (LAG) data structure to develop the image-compressing algorithm, which compresses image file with great efficiency. The rectifying algorithms based on centimeter grid of coordinates, which perfectly rectify the distortion of parameter graph, are proposed. The experimental results show that the PCTR approach in the system can work effectively.
Published in:
Machine Learning and Cybernetics, 2009 International Conference on
(Volume:3
)
Date of Conference: 12-15 July 2009