博客
关于我
强烈建议你试试无所不能的chatGPT,快点击我
使用贝塞尔曲线进行插值 一种非常简单的平滑多边形的方法
阅读量:6874 次
发布时间:2019-06-26

本文共 6324 字,大约阅读时间需要 21 分钟。

原文

Initially, there was a question in  how to interpolate a polygon with a curve in such a way that the resulting curve would be smooth and hit all its vertices. suggested to use a well-known B-Spline interpolation. . B-Spline works good and it behaves like an elastic ruler fixed in the polygon vertices.

 

 

But I had a gut feeling that there must be a simpler method. For example, approximation with cubic Bezier curves. A Bezier curve has two anchor points (begin and end) and two control ones (CP) that determine its shape. More information about Bezier curves can be found using any , for example, on . Our anchor points are given, they are pair of vertices of the polygon. The question was, how to calculate the control points. I ran  and drew this picture. It was pretty easy and I decided to try to calculate their coordinates. It was obvious that the control points of two adjacent edges plus the vertex between them should form one straight line. Only in this case the two adjacent curves will be connected smoothly. So, the two CP should be the a reflection of each other, but… not quite. Reflection assumes equal distances from the central point. For our case it's not correct. First, I tried to calculate a  between two edges and then take points on the perpendicular to it. But as shown in the picture, the CP not always lie on the perpendicular to the .

 

Finally, I found a very simple method that does not require any complicated math. First, we take the polygon and calculate the middle points Ai of its edges.

 

Here we have line segments Ci that connect two points Ai of the adjacent segments. Then, we should calculate points Bias shown in this picture.

 

The third step is final. We simply move the line segments Ci in such a way that their points Bi coincide with the respective vertices. That's it, we calculated the control points for our Bezier curve and the result looks good.

 

One little improvement. Since we have a straight line that determines the place of our control points, we can move them as we want, changing the shape of the resulting curve. I used a simple coefficient K that moves the points along the line relatively to the initial distance between vertices and control points. The closer the control points to the vertices are the sharper figure will be obtained.

 

Below is the result of rendering a popular in  lion in its original form and with Bezier interpolation with K=1.0

 

 

And the enlarged ones.

 

 

The method works quite well with self-intersecting polygons. The examples below show that the result is pretty interesting.

 
 
 

This method is pure heuristic and empiric. It probably gives a wrong result from the point of view of strict mathematical modeling. But in practice the result is good enough and it requires absolute minimum of calculations. Below is the source code that has been used to generate the lions shown above. It's not optimal and just an illustration. It calculates some variables twice, while in real programs we can store and reuse them in the consecutive steps.

// Assume we need to calculate the control    // points between (x1,y1) and (x2,y2).    // Then x0,y0 - the previous vertex,    //      x3,y3 - the next one.    double xc1 = (x0 + x1) / 2.0; double yc1 = (y0 + y1) / 2.0; double xc2 = (x1 + x2) / 2.0; double yc2 = (y1 + y2) / 2.0; double xc3 = (x2 + x3) / 2.0; double yc3 = (y2 + y3) / 2.0; double len1 = sqrt((x1-x0) * (x1-x0) + (y1-y0) * (y1-y0)); double len2 = sqrt((x2-x1) * (x2-x1) + (y2-y1) * (y2-y1)); double len3 = sqrt((x3-x2) * (x3-x2) + (y3-y2) * (y3-y2)); double k1 = len1 / (len1 + len2); double k2 = len2 / (len2 + len3); double xm1 = xc1 + (xc2 - xc1) * k1; double ym1 = yc1 + (yc2 - yc1) * k1; double xm2 = xc2 + (xc3 - xc2) * k2; double ym2 = yc2 + (yc3 - yc2) * k2; // Resulting control points. Here smooth_value is mentioned // above coefficient K whose value should be in range [0...1]. ctrl1_x = xm1 + (xc2 - xm1) * smooth_value + x1 - xm1; ctrl1_y = ym1 + (yc2 - ym1) * smooth_value + y1 - ym1; ctrl2_x = xm2 + (xc2 - xm2) * smooth_value + x2 - xm2; ctrl2_y = ym2 + (yc2 - ym2) * smooth_value + y2 - ym2;

And the source code of an approximation with a cubic Bezier curve.

// Number of intermediate points between two source ones,// Actually, this value should be calculated in some way,// Obviously, depending on the real length of the curve.// But I don't know any elegant and fast solution for this// problem.#define NUM_STEPS 20 void (Polygon* p, double x1, double y1, //Anchor1 double x2, double y2, //Control1 double x3, double y3, //Control2 double x4, double y4) //Anchor2 { double dx1 = x2 - x1; double dy1 = y2 - y1; double dx2 = x3 - x2; double dy2 = y3 - y2; double dx3 = x4 - x3; double dy3 = y4 - y3; double subdiv_step = 1.0 / (NUM_STEPS + 1); double subdiv_step2 = subdiv_step*subdiv_step; double subdiv_step3 = subdiv_step*subdiv_step*subdiv_step; double pre1 = 3.0 * subdiv_step; double pre2 = 3.0 * subdiv_step2; double pre4 = 6.0 * subdiv_step2; double pre5 = 6.0 * subdiv_step3; double tmp1x = x1 - x2 * 2.0 + x3; double tmp1y = y1 - y2 * 2.0 + y3; double tmp2x = (x2 - x3)*3.0 - x1 + x4; double tmp2y = (y2 - y3)*3.0 - y1 + y4; double fx = x1; double fy = y1; double dfx = (x2 - x1)*pre1 + tmp1x*pre2 + tmp2x*subdiv_step3; double dfy = (y2 - y1)*pre1 + tmp1y*pre2 + tmp2y*subdiv_step3; double ddfx = tmp1x*pre4 + tmp2x*pre5; double ddfy = tmp1y*pre4 + tmp2y*pre5; double dddfx = tmp2x*pre5; double dddfy = tmp2y*pre5; int step = NUM_STEPS; // Suppose, we have some abstract object Polygon which // has method AddVertex(x, y), similar to LineTo in // many graphical APIs. // Note, that the loop has only operation add! while(step--) { fx += dfx; fy += dfy; dfx += ddfx; dfy += ddfy; ddfx += dddfx; ddfy += dddfy; p->AddVertex(fx, fy); } p->AddVertex(x4, y4); // Last step must go exactly to x4, y4 }

You can download a working application for Windows that renders the lion, rotates and scales it, and generates random polygons. . Press left mouse button and drag to rotate and scale the image around the center point. Press right mouse button and drag left-right to change the coefficient of smoothing (K). Value K=1 is about 100 pixels from the left border of the window. Each left double-click generates a random polygon. You can also rotate and scale it, and change K.

 
Copyright © 2002-2006 
Web Design and Programming 

转载地址:http://sbpfl.baihongyu.com/

你可能感兴趣的文章
优化Android Studio
查看>>
zabbix二次开发-flask-获取告警
查看>>
我的友情链接
查看>>
java实现MD5加密处理
查看>>
实用JVM参数总结
查看>>
oracle 11g R2 64位 安装详细步骤
查看>>
Jpeg 库的解码OpenCL优化
查看>>
正则表达式
查看>>
『中级篇』docker之虚拟机创建vagrant技巧(番外篇)(81)
查看>>
交换机SPAN功能配置
查看>>
MySQL 架构组成—存储引擎
查看>>
基于数值分析思想对多项式求值的原理和应用进行探究
查看>>
vue-devtools vue开发调试神器
查看>>
PHP扩展模块的安装
查看>>
BGP基础操作
查看>>
SimpleXml项目
查看>>
php下使用PDO创建sqlite3数据库
查看>>
Istio技术与实践6:Istio如何为服务提供安全防护能力
查看>>
ISTP的重要作用
查看>>
驼峰设计 PPT美化
查看>>