A scanning image vectorization system was developed in combination of characteristic of big scale scanning relief map.
SIMD optimization is very similar to traditional automatic vectorization for vector processors.
We use a vectorization algorithm to extract form lines from run length connect graph which is calculated first, and in the same time the skew Angle is detected.
According to the features, we finish the vectorization of texts, and then use support vector machine classifier to distinguish texts and filter illegal texts.
An effective method of vectorization in which both raster graphics and image data are converted into polygon features through tracing the run-length outline is developed.
This is only a simple example, demonstrating the basic idea behind auto-vectorization.
Conditions for vectorization of FORTRAN DO-loops are discussed in this paper. Based on these conditions, the decision theorems and the algorithm for vectorization are given.
