一、代码

Python

import numpy as np
import open3d as o3d
from scipy.spatial import cKDTree
from joblib import Parallel, delayeddef calculate_response(idx, points, radius, t, kdtree):# 使用KD树找到半径内的所有点neighbor_indices = kdtree.query_ball_point(points[idx], radius)neighbor_points = points[neighbor_indices]# 计算距离distances = np.linalg.norm(neighbor_points - points[idx], axis=1)# 计算响应值response = np.sum(distances < t) / len(neighbor_indices)return response if response > 0 else 0, idxdef susan_keypoint_detection_optimized(pcd, radius=0.05, t=0.01, response_threshold=0.5, n_jobs=-1):points = np.asarray(pcd.points)kdtree = cKDTree(points)# 并行计算每个点的响应值results = Parallel(n_jobs=n_jobs)(delayed(calculate_response)(i, points, radius, t, kdtree) for i in range(len(points)))# 分离响应值和索引responses, indices &