Point Cloud Minimum Spanning Tree (MST) Calculation using Kruskal's Algorithm

This code snippet calculates the minimum spanning tree (MST) of a point cloud using Kruskal's algorithm.

First, the code loads a point cloud from a PLY file and computes the centroid and normals of the point cloud.

Then, it creates a graph with V vertices and E edges, where V is the number of points in the point cloud and E is the total number of possible edges.

Next, it calculates the edge weights based on the Euclidean distance between points and adds these edges to the graph.

Finally, it executes the Kruskal's algorithm to find the MST of the graph. The result is stored in the 'result' vector.

int main()
{
	std::vector<Edge> edges;
	pcl::PointCloud<pcl::PointXYZ>::Ptr singlejiedian(new pcl::PointCloud<pcl::PointXYZ>);
	// Load the input point cloud from PLY file
	pcl::PointCloud<pcl::PointXYZ>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZ>);
	pcl::io::loadPLYFile<pcl::PointXYZ>('D:\DIANYUNWENJIANJIA\newOUSHIJULEI_ply.ply', *cloud);

	// Compute the centroid of the point cloud
	Eigen::Vector4f centroid;
	pcl::compute3DCentroid(*cloud, centroid);

	// Compute the normals of the point cloud
	pcl::NormalEstimation<pcl::PointXYZ, pcl::Normal> ne;
	pcl::PointCloud<pcl::Normal>::Ptr cloud_normals(new pcl::PointCloud<pcl::Normal>);
	pcl::search::KdTree<pcl::PointXYZ>::Ptr tree(new pcl::search::KdTree<pcl::PointXYZ>);
	ne.setInputCloud(cloud);
	ne.setSearchMethod(tree);
	ne.setKSearch(40);
	ne.compute(*cloud_normals);
	// Create a graph with V vertices and E edges
	int V = cloud->size();
	int E = V * (V - 1) / 2;
	Graph graph(V, E);
	// Calculate the edge weights based on Euclidean distance between points
	for (int i = 0; i < V - 1; ++i)
	{
		const auto& src_point = cloud->points[i];
		for (int j = i + 1; j < V; ++j)
		{
			const auto& tgt_point = cloud->points[j];
			float distance = euclideanDistance(src_point, tgt_point);
			graph.addEdge(i, j, distance);
		}
	}
	// 执行最小生成树
	std::vector<Edge> result;
	graph.KruskalMST(cloud, result);
	//给这段代码加上，pcl1.8.1可以运行的功能，先把这些质心点的曲率算出来，然后做一个曲率筛选，筛选出叶片和茎干，用这些点再做最小生成树
}

To enhance this code, you can incorporate the following functionalities using PCL 1.8.1:

1. Calculate Curvature: Compute the curvature of the centroid points. You can utilize the pcl::SHOTEstimation class for this purpose. This class computes the Shape Context descriptor, which includes curvature information.

2. Curvature Filtering: Implement a curvature-based filter to select points that represent leaves and stems. You can define thresholds for curvature values to distinguish between these features.

3. MST with Filtered Points: Use the filtered points to construct a new graph and then apply Kruskal's algorithm to find the MST. This will create an MST based on the selected leaf and stem points.

By incorporating these features, the code will be more sophisticated and capable of identifying and extracting the desired leaf and stem structures from the point cloud.