Wu的抗锯齿算法，垂直线
我主要按照 Xiaolin Wu 的算法 实现 Wu 的算法，但是遇到了有点麻烦。具体来说，该算法的这一点包含在 wiki 条目底部的注释中：
如果在例程开始时abs(dx) <绝对值（dy）为真， 那么所有的绘图都应该在 x 和 y 颠倒的情况下完成。
我认为这意味着只需用plot(y, x)反转对plot(x, y)的所有调用，但这样做会导致一些非常特殊的线条（我似乎无法获得屏幕截图，因为每次我尝试，我的 OpenGL 窗口将空白粘贴到 Paint 中）。
以前实施过此操作的人可以给我一些指导吗？我的线条现在看起来有点傻，每个象限只填充了一半。
I've been implementing Wu's algorithm mostly per Xiaolin Wu's algorithm, but have run into a bit of a snag. Specifically, this bit of the algorithm, which is included by a note at the bottom of the wiki entry:
If at the beginning of the routine abs(dx) < abs(dy) is true, then all plotting should be done with x and y reversed.
I thought that this meant just reverse all the calls to plot(x, y) with plot(y, x), but doing so resulted in some very special looking lines (I can't seem to get a screenshot because every time I try, my OpenGL window pastes blank into Paint).
Can anyone who has implemented this before give me a bit of guidance? My lines look a bit silly right now with only half of each quadrant filled.
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
您不仅应该将
plot(x,y)与plot(y,x)交换，还应该交换输入参数x1<->y1> 和x2<->y2。如果您同时执行这两项操作，它将正常工作。后一部分是在文章的代码中完成的，但没有进行绘图坐标的切换。背后的原因是您在
x范围内逐像素步进。如果您的水平范围小于垂直范围，则可能会产生间隙（例如，一条完全垂直的线将导致它仅产生一个 x 值）。因此，您可以切换输入参数
x和y，但同时切换输出坐标系（通过交换x和y> 在绘图函数中）。You should not only swap
plot(x,y)withplot(y,x)but also exchange the input parametersx1<->y1andx2<->y2. If you do both it´ll work correctly. The latter part is done in the code in the article but not the switch of the plot coordinates.Reason behind is that you step pixel by pixel in
xrange. If your horizontal extent is less than vertical it might yield gaps (think e.g. a perfectly vertical line which would result it a single x-value only).Therefore you switch input parameters
xandybut at the same time the output coordinate system (by exchangingxandyin the plot function).