Python如何生成树形图案
作者:oldjwu 时间:2022-11-17 03:39:58
本文实例为大家分享了Python生成树形图案的具体代码,供大家参考,具体内容如下
先看一下效果,见下图。
上面这颗大树是使用Python + Tkinter绘制的,主要原理为使用分形画树干、树枝,最终叶节点上画上绿色圆圈代表树叶。当然,为了看起来更真实,绘制过程中也加入了一些随机变化,比如树枝会稍微有些扭曲而不是一条直线,分叉的角度、长短等都会随机地作一些偏移等。
以下是完整源代码:
# -*- coding: utf-8 -*-
import Tkinter
import sys, random, math
class Point(object):
def __init__(self, x, y):
self.x = x
self.y = y
def __str__(self):
return "<Point>: (%f, %f)" % (self.x, self.y)
class Branch(object):
def __init__(self, bottom, top, branches, level = 0):
self.bottom = bottom
self.top = top
self.level = level
self.branches = branches
self.children = []
def __str__(self):
s = "Top: %s, Bottom: %s, Children Count: %d" % /
(self.top, self.bottom, len(self.children))
return s
def nextGen(self, n = -1, rnd = 1):
if n <= 0: n = self.branches
if rnd == 1:
n = random.randint(n / 2, n * 2)
if n <= 0: n = 1
dx = self.top.x - self.bottom.x
dy = self.top.y - self.bottom.y
r = 0.20 + random.random() * 0.2
if self.top.x == self.bottom.x:
# 如果是一条竖线
x = self.top.x
y = dy * r + self.bottom.y
elif self.top.y == self.bottom.y:
# 如果是一条横线
x = dx * r + self.bottom.x
y = self.top.y
else:
x = dx * r
y = x * dy / dx
x += self.bottom.x
y += self.bottom.y
oldTop = self.top
self.top = Point(x, y)
a = math.pi / (2 * n)
for i in range(n):
a2 = -a * (n - 1) / 2 + a * i - math.pi
a2 *= 0.9 + random.random() * 0.2
self.children.append(self.mkNewBranch(self.top, oldTop, a2))
def mkNewBranch(self, bottom, top, a):
dx1 = top.x - bottom.x
dy1 = top.y - bottom.y
r = 0.9 + random.random() * 0.2
c = math.sqrt(dx1 ** 2 + dy1 ** 2) * r
if dx1 == 0:
a2 = math.pi / 2
else:
a2 = math.atan(dy1 / dx1)
if (a2 < 0 and bottom.y > top.y) /
or (a2 > 0 and bottom.y < top.y) /
:
a2 += math.pi
b = a2 - a
dx2 = c * math.cos(b)
dy2 = c * math.sin(b)
newTop = Point(dx2 + bottom.x, dy2 + bottom.y)
return Branch(bottom, newTop, self.branches, self.level + 1)
class Tree(object):
def __init__(self, root, canvas, bottom, top, branches = 3, depth = 3):
self.root = root
self.canvas = canvas
self.bottom = bottom
self.top = top
self.branches = branches
self.depth = depth
self.new()
def gen(self, n = 1):
for i in range(n):
self.getLeaves()
for node in self.leaves:
node.nextGen()
self.show()
def new(self):
self.leavesCount = 0
self.branch = Branch(self.bottom, self.top, self.branches)
self.gen(self.depth)
print "leaves count: %d" % self.leavesCount
def chgDepth(self, d):
self.depth += d
if self.depth < 0: self.depth = 0
if self.depth > 10: self.depth = 10
self.new()
def chgBranch(self, d):
self.branches += d
if self.branches < 1: self.branches = 1
if self.branches > 10: self.branches = 10
self.new()
def getLeaves(self):
self.leaves = []
self.map(self.findLeaf)
def findLeaf(self, node):
if len(node.children) == 0:
self.leaves.append(node)
def show(self):
for i in self.canvas.find_all():
self.canvas.delete(i)
self.map(self.drawNode)
self.canvas.tag_raise("leaf")
def exit(self, evt):
sys.exit(0)
def map(self, func = lambda node: node):
# 遍历树
children = [self.branch]
while len(children) != 0:
newChildren = []
for node in children:
func(node)
newChildren.extend(node.children)
children = newChildren
def drawNode(self, node):
self.line2(
# self.canvas.create_line(
node.bottom.x,
node.bottom.y,
node.top.x,
node.top.y,
fill = "#100",
width = 1.5 ** (self.depth - node.level),
tags = "branch level_%d" % node.level,
)
if len(node.children) == 0:
# 画叶子
self.leavesCount += 1
self.canvas.create_oval(
node.top.x - 3,
node.top.y - 3,
node.top.x + 3,
node.top.y + 3,
fill = "#090",
tag = "leaf",
)
self.canvas.update()
def line2(self, x0, y0, x1, y1, width = 1, fill = "#000", minDist = 10, tags = ""):
dots = midDots(x0, y0, x1, y1, minDist)
dots2 = []
for i in range(len(dots) - 1):
dots2.extend([dots[i].x,
dots[i].y,
dots[i + 1].x,
dots[i + 1].y])
self.canvas.create_line(
dots2,
fill = fill,
width = width,
smooth = True,
tags = tags,
)
def midDots(x0, y0, x1, y1, d):
dots = []
dx, dy, r = x1 - x0, y1 - y0, 0
if dx != 0:
r = float(dy) / dx
c = math.sqrt(dx ** 2 + dy ** 2)
n = int(c / d) + 1
for i in range(n):
if dx != 0:
x = dx * i / n
y = x * r
else:
x = dx
y = dy * i / n
if i > 0:
x += d * (0.5 - random.random()) * 0.25
y += d * (0.5 - random.random()) * 0.25
x += x0
y += y0
dots.append(Point(x, y))
dots.append(Point(x1, y1))
return dots
if __name__ == "__main__":
root = Tkinter.Tk()
root.title("Tree")
gw, gh = 800, 600
canvas = Tkinter.Canvas(root,
width = gw,
height = gh,
)
canvas.pack()
tree = Tree(root, canvas, Point(gw / 2, gh - 20), Point(gw / 2, gh * 0.2), /
branches = 2, depth = 8)
root.bind("n", lambda evt: tree.new())
root.bind("=", lambda evt: tree.chgDepth(1))
root.bind("+", lambda evt: tree.chgDepth(1))
root.bind("-", lambda evt: tree.chgDepth(-1))
root.bind("b", lambda evt: tree.chgBranch(1))
root.bind("c", lambda evt: tree.chgBranch(-1))
root.bind("q", tree.exit)
root.mainloop()
因为每次生成的树都是随机的,所以你生成的树和上图会不太一样,可能会更为枝繁叶茂,也可能会看起来才刚刚发芽。程序中绑定了若干快捷键,比如“n”是随机产生一颗新的树,“q”是退出程序。另外还有一些不太常用的快捷键,如“+”/“-”是增加/减少树的深度,“b”/“c”分别代表更多/更少的分叉,需要注意的是,增加深度或分叉可能需要更多的计算时间。
从这次树形图案的绘制过程中,我也有一些有趣的发现,比如,树枝上某一处的横截面宽度与它与树根之间的距离似乎呈一种指数函数的关系。如用H表示树的总高度,h表示树枝上某一点的高度,w表示这一点横截面的宽度,那么w与h之间似乎存在这样一种关系:w = a * b ^ (H - h) + c,这儿a、b、c都是常数。当然,这只是一个猜测,因为绘制的过程中我发现当w与h满足这样关系时画出来的图案看起来最“自然”,这个问题或许下次可以再深入研究一下。
来源:http://blog.csdn.net/oldjwu/article/details/4651146
标签:Python,树形图案
0
投稿
猜你喜欢
关于numpy中np.nonzero()函数用法的详解
2023-01-06 12:29:34
Git 教程之标签详解
2023-10-25 21:58:04
Python中的测试模块unittest和doctest的使用教程
2021-09-07 05:50:54
PHP解决高并发问题(opcache)
2023-11-07 23:17:49
对python 自定义协议的方法详解
2023-05-26 07:20:04
Python使用chardet判断字符编码
2021-05-14 03:03:52
javascript学习随笔(编写浏览器脚本 Navigator Scripting )
2024-04-28 09:49:39
Node.js中的process.nextTick使用实例
2024-05-08 09:36:18
详解 Mysql 事务和Mysql 日志
2024-01-19 05:59:17
VUEJS实战之构建基础并渲染出列表(1)
2024-05-29 22:14:46
vue使用Element el-upload 组件踩坑记
2023-07-02 16:32:17
python实现简单飞机大战小游戏
2023-12-19 06:55:37
详解Python中的__getitem__方法与slice对象的切片操作
2022-04-26 19:54:55
OpenCV-Python实现油画效果的实例
2022-03-26 09:16:37
python网络爬虫精解之正则表达式的使用说明
2023-02-26 17:13:21
js实现滑动进度条效果
2023-08-24 03:47:39
详解 Python 读写XML文件的实例
2022-05-03 23:36:15
iscroll动态加载数据完美解决方法
2024-04-10 13:59:01
python 贪心算法的实现
2023-01-25 08:18:50
Python编程源码报错解决方法总结经验分享
2023-11-16 07:49:46