python练习9 Open3D处理

一、open3d的安装与介绍 1.安装open3d

pip install open3d 或 pip3 install open3d

查看安装包：pip list

2.open3d简介 

---

二.案例：斯坦福兔子 1.生成点云

①不同角度扫描到的兔子的点云

以bun000为例, 可以看出生成的是一片，不是整个兔子

 

import open3d as o3d
import numpy as np
import os
print(os.getcwd())

print("Open3D read Point Cloud")
pcd=o3d.io.read_point_cloud(r"D:pythonAds2021bunnydatabun000.ply")
print(pcd)

o3d.visualization.draw_geometries([pcd],width=800,height=600)

 ②多角度点云拼在一起生成整个点云（有大量重复）

import open3d as o3d
import numpy as np
import os
print(os.getcwd())

print("Open3D read Point Cloud")
pcd=o3d.io.read_point_cloud(r"D:pythonAds2021bunnybunny10k.ply")
print(pcd)

o3d.visualization.draw_geometries([pcd],width=800,height=600)

按ctrl+-号可以减少点云的体素尺寸，ctrl+放大，ctrl-缩小

 看看open3d提供了哪些功能

 2.近邻搜索

文档：http://www.open3d.org/docs/release/

①周围n个点

import open3d as o3d
import numpy as np

print("Open3D read Point Cloud")
pcd = o3d.io.read_point_cloud(r"D:pythonAds2021bunnybunny10k.ply")
pcd.paint_uniform_color([0.5, 0.5, 0.5])#这里的rgb取值为0-1，[0.5,0.5,0.5]意味着整个兔子的点云都以灰色呈现

pcd_tree = o3d.geometry.KDTreeFlann(pcd)#KD树贴着物体表面去找近邻的点
pcd.colors[100] = [1, 0, 0]#兔子点云的第100个点以红色呈现（一共6K+个点，不能超过这个范围）

[k, idx, _] = pcd_tree.search_knn_vector_3d(pcd.points[100],100)#对树做近邻搜索，以第100个点为中心，找附近的100个点
np.asarray(pcd.colors)[idx[1:], :] = [0, 1, 0]#将返回找到的点的坐标放到一个数组里，色彩设置为绿色

o3d.visualization.draw_geometries([pcd],width=1200,height=1000)#绘制pcd

 ②索引半径

[k,idx,_] = pcd_tree.search_radius_vector_3d(pcd.points[3000],0.1) #索引半径小于0.02
np.asarray(pcd.colors)[idx[1:], :] = [0, 0, 1]

 ③混合搜索

pcd.colors[2000]=[1, 0, 0]
[k2, idx2, _]=pcd_tree.search_hybrid_vector_3d(pcd.points[2000],0.05,200)
np.asarray(pcd.colors)[idx2[1:], :] = [0, 1, 0.8]
o3d.visualization.draw_geometries([pcd],width=1200,height=1000)

 3.法向量估计

令点与其他临近点生成平面，根据距平面最近距离判断法向量

import open3d as o3d
import numpy as np
 
print("Open3D read Point Cloud")
pcd = o3d.io.read_point_cloud(r"D:pythonAds2021bunnybunny10k.ply")
print(pcd)
dumppcd = pcd.voxel_down_sample(voxel_size=0.01)  #下采样（降采样）
 
dumppcd.estimate_normals(search_param=o3d.geometry.KDTreeSearchParamHybrid(radius=0.01,max_nn=30))
 
print(dumppcd.normals[0])
print(np.asarray(dumppcd.normals)[:10,:])
 
o3d.visualization.draw_geometries([dumppcd],point_show_normal=True,
                                  window_name="法线估计", width=1200,height=1000,                                mesh_show_back_face=False)

 4.用无结构的点云数据生成结构化数据Mesh

参考:https://blog.csdn.net/io569417668/article/details/110855147

 

 5.用三角片面生成结构化数据Mesh

生成三角片面先做法向量估计，数据量更小，判断哪些点应该相连生成平面

存了面的信息，面都是三角形的

①mesh模型

创建球体，生成点云信息，点数越少越不圆

 REF算法：http://www.open3d.org/docs/0.10.0/tutorial/reference.html#Bernardini1999

mesh = o3d.geometry.TriangleMesh.create_sphere()
mesh.compute_vertex_normals()
o3d.visualization.draw_geometries([mesh])
pcd = mesh.sample_points_uniformly(number_of_points=500)
o3d.visualization.draw_geometries([pcd])

 ②兔子案例

import open3d as o3d
# import open3d_tutorial as o3dtut
import numpy as np

print("Open3D read Point Cloud")
pcd = o3d.io.read_triangle_mesh(r"D:pythonAds2021bunnybunny10k.ply")  #newrabbit.pcd")
print(pcd)
pcd.compute_vertex_normals()
pcdmesh = pcd.sample_points_poisson_disk(3000)
o3d.visualization.draw_geometries([pcdmesh],point_show_normal=True)

radii=[0.005, 0.01, 0.02, 0.04]

ballmesh = o3d.geometry.TriangleMesh.create_from_point_cloud_ball_pivoting(pcdmesh,o3d.utility.DoubleVector(radii))
print(ballmesh)
o3d.visualization.draw_geometries([ballmesh])

o3d.visualization.draw_geometries([pcd, ballmesh])