注意那些使用potrace和imagemagick的人，将透明的PNG图像转换为PPM似乎不是很好。下面是一个在convert上使用-flatten标志来处理这个问题的例子:

sudo apt-get install potrace imagemagick
convert -flatten input.png output.ppm
potrace -s output.ppm -o output.svg
rm output.ppm

Another interesting phenomenon is that you can use PPM (256*3 colors, ie. RGB), PGM (256 colors, ie. grayscale) or PBM (2 colors, ie. white or black only) as the input format. From my limited observations, it would appear that on images which are anti-aliased, PPM and PGM (which produce identical SVGs as far as I can see) shrink the colored area and PBM expands the colored area (albeit only a little). Presumably this is the difference between a pixel > (256 / 2) test and a pixel > 0 test. You can switch between the three by changing the file extension, ie. the following use PBM:

sudo apt-get install potrace imagemagick
convert -flatten input.png output.pbm
potrace -s output.pbm -o output.svg
rm output.pbm

你可能想看看potrace。

png是位图图像风格，SVG是支持位图的基于矢量的图形设计，因此它不会将图像转换为矢量，只是将图像嵌入到基于矢量的格式中。你可以使用免费的http://www.inkscape.org/来实现。它会嵌入它，但是它也有一个Live Trace之类的引擎，如果你愿意，它会尝试将它转换为路径(使用potrace)。查看adobe illustrator中的实时跟踪(商业)是一个例子:

http://graphicssoft.about.com/od/illustrator/ss/sflivetrace.htm

你也可以试试http://image.online-convert.com/convert-to-svg

我总是用它来满足我的需要。

这个工具现在工作得很好。

http://www.mobilefish.com/services/image2svg/image2svg.php

I'm assuming that you wish to write software to do this. To do it naively you would just find lines and set the vectors. To do it intelligently, you attempt to fit shapes onto the drawing (model fitting). Additionally, you should attempt to ascertain bitmaped regions (regions you can't model through shames or applying textures. I would not recommend going this route as that it will take quite a bit of time and require a bit of graphics and computer vision knowledge. However, the output will much and scale much better than your original output.