map_evaluation_tools

This tool computes the Mean Map Entropy and the Mean Plane Variance of a point cloud.

##Dependencies

• PCL
• OpenMP
• Eigen
• pybind11

Install

1. First, clone this project，notice the SUBMODULE should also be cloned.

git clone https://github.com/hyxhope/map_evaluation_tools.git --recursive

2. Compile and install.

python setup.py install

Usage

For details, please refer scripts/example.py

Example

import numpy as np
import map_evaluation

# 从pcd文件加载点云
pcd_file = "your/path/to/pointcloud.pcd"
map_evaluation.init(pcd_file=pcd_file, voxelLeaf=0.3, minKNearest=5, stepSize=2, radius=1.0)

# map_evaluation同时支持numpy传入N*4点云
pc = np.empty((25000, 4)) # x,y,z,intensity
map_evaluation.init(input=pc, voxelLeaf=0.3, minKNearest=5, stepSize=2, radius=1.0)

# mme, mpv
print(map_evaluation.mme())
print(map_evaluation.mpv())

Parameter

// 参数设置(Optional)
float voxelLeaf;    // 降采样阈值(可以为0，表示不进行降采样操作)
int minKNearest;    // 最小最近邻邻居的个数
int stepSize;       // 步长(每隔几个点计算一次，可加快计算速度)
float radius;       // 最近邻搜索阈值

TODO

Add mom