无纸动漫领域中位图矢量化的算法研究

Algorithmic Study on the Vectorization of Bitmap in Paperless Cartoon Filed

位图图像的矢量化是计算机图形学的热点问题。图像的矢量化解决了速度和空间上的问题,为图像进一步的输出与管理提供了便利。因此,他也是无纸动漫领域所面临的基础问题之一。通过对位图矢量化课题的深入研究,针对矢量化过程中的几个关键点技术,特别针对图像轮廓的提取,跟踪及关键特征点提取和曲线拟合,提出行之有效的解决算法。最后在VC 6.0环境下对算法进行实现。其处理速度快、曲线拟合光滑、失真小,验证了算法的可行性,为其在一个无纸动漫系统中的初步应用奠定了基础。

Vectorization of bitmap is a hot issue of the computer graphics. It not only resolves problems of the velocity and space, but also provides some advantages for the further output and the management of images. Therefore, it is also one of the basic problems facing the paperless cartoon. Thorough the in- depth study of the vectorization of Bitmap, this paper puts forward the effective algorithm of myself which aims at the several key technical points,especially in the extraction of image contour, the contour tracking, the extraction of key characteristic points and the curve fitting. Finally, the arithmetic has been programmed based on Visual C＋＋ 6.0 and been applied in a system of paperless cartoon. It proves the high speed and feasibility because he curve is smooth and the distortion of the image is very small. It establishes the foundation for the elementary application in a paperless cartoon system.