first commit

This commit is contained in:
龙澳 2024-12-23 11:31:20 +08:00
commit 17deadb25e
16 changed files with 1774 additions and 0 deletions

7
.gitignore vendored Normal file
View File

@ -0,0 +1,7 @@
# 忽略所有__pycache__目录
**/__pycache__/
*.pyc
*.pyo
*.pyd
test/

2
README.md Normal file
View File

@ -0,0 +1,2 @@
# ODM_Pro
无人机三维重建

82
filter/cluster_filter.py Normal file
View File

@ -0,0 +1,82 @@
from sklearn.cluster import DBSCAN
from sklearn.preprocessing import StandardScaler
import os
class GPSCluster:
def __init__(self, gps_points, output_dir: str, eps=0.01, min_samples=5):
"""
初始化GPS聚类器
参数:
eps: DBSCAN的邻域半径参数
min_samples: DBSCAN的最小样本数参数
"""
self.eps = eps
self.min_samples = min_samples
self.dbscan = DBSCAN(eps=eps, min_samples=min_samples)
self.scaler = StandardScaler()
self.gps_points = gps_points
def fit(self):
"""
对GPS点进行聚类只保留最大的类
参数:
gps_points: 包含'lat''lon'列的DataFrame
返回:
带有聚类标签的DataFrame其中最大类标记为1其他点标记为-1
"""
# 提取经纬度数据
X = self.gps_points[["lon", "lat"]].values
# # 数据标准化
# X_scaled = self.scaler.fit_transform(X)
# 执行DBSCAN聚类
labels = self.dbscan.fit_predict(X)
# 找出最大类的标签(排除噪声点-1
unique_labels = [l for l in set(labels) if l != -1]
if unique_labels: # 如果有聚类
label_counts = [(l, sum(labels == l)) for l in unique_labels]
largest_label = max(label_counts, key=lambda x: x[1])[0]
# 将最大类标记为1其他都标记为-1
new_labels = (labels == largest_label).astype(int)
new_labels[new_labels == 0] = -1
else: # 如果没有聚类,全部标记为-1
new_labels = labels
# 将聚类结果添加到原始数据中
result_df = self.gps_points.copy()
result_df["cluster"] = new_labels
return result_df
def get_cluster_stats(self, clustered_points):
"""
获取聚类统计信息
参数:
clustered_points: 带有聚类标签的DataFrame
返回:
聚类统计信息的字典
"""
main_cluster_points = sum(clustered_points["cluster"] == 1)
stats = {
"total_points": len(clustered_points),
"main_cluster_points": main_cluster_points,
"noise_points": sum(clustered_points["cluster"] == -1),
}
noise_cluster = self.get_noise_cluster(clustered_points)
return stats
def get_main_cluster(self, clustered_points):
return clustered_points[clustered_points["cluster"] == 1]
def get_noise_cluster(self, clustered_points):
return clustered_points[clustered_points["cluster"] == -1]

248
filter/gps_filter.py Normal file
View File

@ -0,0 +1,248 @@
import os
import math
from itertools import combinations
import numpy as np
from scipy.spatial import KDTree
import logging
import pandas as pd
from datetime import datetime, timedelta
class GPSFilter:
"""过滤密集点及孤立点"""
def __init__(self, output_dir):
self.logger = logging.getLogger('UAV_Preprocess.GPSFilter')
@staticmethod
def _haversine(lat1, lon1, lat2, lon2):
"""计算两点之间的地理距离(单位:米)"""
R = 6371000 # 地球平均半径,单位:米
phi1, phi2 = math.radians(lat1), math.radians(lat2)
delta_phi = math.radians(lat2 - lat1)
delta_lambda = math.radians(lon2 - lon1)
a = math.sin(delta_phi / 2) ** 2 + math.cos(phi1) * \
math.cos(phi2) * math.sin(delta_lambda / 2) ** 2
c = 2 * math.atan2(math.sqrt(a), math.sqrt(1 - a))
return R * c
@staticmethod
def _assign_to_grid(lat, lon, grid_size, min_lat, min_lon):
"""根据经纬度和网格大小,将点分配到网格"""
grid_x = int((lat - min_lat) // grid_size)
grid_y = int((lon - min_lon) // grid_size)
return grid_x, grid_y
def _get_distances(self, points_df, grid_size):
"""读取图片 GPS 坐标,计算点对之间的距离并排序"""
# 确定经纬度范围
min_lat, max_lat = points_df['lat'].min(), points_df['lat'].max()
min_lon, max_lon = points_df['lon'].min(), points_df['lon'].max()
self.logger.info(
f"经纬度范围:纬度[{min_lat:.6f}, {max_lat:.6f}],纬度范围[{max_lat-min_lat:.6f}]"
f"经度[{min_lon:.6f}, {max_lon:.6f}],经度范围[{max_lon-min_lon:.6f}]")
# 分配到网格
grid_map = {}
for _, row in points_df.iterrows():
grid = self._assign_to_grid(
row['lat'], row['lon'], grid_size, min_lat, min_lon)
if grid not in grid_map:
grid_map[grid] = []
grid_map[grid].append((row['file'], row['lat'], row['lon']))
self.logger.info(f"图像点已分配到 {len(grid_map)} 个网格中")
# 在每个网格中计算两两距离并排序
sorted_distances = {}
for grid, images in grid_map.items():
distances = []
for (img1, lat1, lon1), (img2, lat2, lon2) in combinations(images, 2):
dist = self._haversine(lat1, lon1, lat2, lon2)
distances.append((img1, img2, dist))
distances.sort(key=lambda x: x[2]) # 按距离升序排序
sorted_distances[grid] = distances
self.logger.debug(f"网格 {grid} 中计算了 {len(distances)} 个距离对")
return sorted_distances
def _group_by_time(self, points_df: pd.DataFrame, time_threshold: timedelta) -> list:
"""根据拍摄时间分组图片
如果相邻两张图片的拍摄时间差超过5分钟则进行切分
Args:
points_df: 包含图片信息的DataFrame必须包含'file''date'
time_threshold: 时间间隔阈值默认5分钟
Returns:
list: 每个元素为时间组内的点数据
"""
if 'date' not in points_df.columns:
self.logger.error("数据中缺少date列")
return [points_df]
# 将date为空的行单独作为一组
null_date_group = points_df[points_df['date'].isna()]
valid_date_points = points_df[points_df['date'].notna()]
if not null_date_group.empty:
self.logger.info(f"发现 {len(null_date_group)} 个无时间戳的点,将作为单独分组")
if valid_date_points.empty:
self.logger.warning("没有有效的时间戳数据")
return [null_date_group] if not null_date_group.empty else []
# 按时间排序
valid_date_points = valid_date_points.sort_values('date')
self.logger.info(
f"有效时间范围: {valid_date_points['date'].min()}{valid_date_points['date'].max()}")
# 计算时间差
time_diffs = valid_date_points['date'].diff()
# 找到时间差超过阈值的位置
time_groups = []
current_group_start = 0
for idx, time_diff in enumerate(time_diffs):
if time_diff and time_diff > time_threshold:
# 添加当前组
current_group = valid_date_points.iloc[current_group_start:idx]
time_groups.append(current_group)
# 记录断点信息
break_time = valid_date_points.iloc[idx]['date']
group_start_time = current_group.iloc[0]['date']
group_end_time = current_group.iloc[-1]['date']
self.logger.info(
f"时间组 {len(time_groups)}: {len(current_group)} 个点, "
f"时间范围 [{group_start_time} - {group_end_time}]"
)
self.logger.info(
f"在时间 {break_time} 处发现断点,时间差为 {time_diff}")
current_group_start = idx
# 添加最后一组
last_group = valid_date_points.iloc[current_group_start:]
if not last_group.empty:
time_groups.append(last_group)
self.logger.info(
f"时间组 {len(time_groups)}: {len(last_group)} 个点, "
f"时间范围 [{last_group.iloc[0]['date']} - {last_group.iloc[-1]['date']}]"
)
# 如果有空时间戳的点,将其作为最后一组
if not null_date_group.empty:
time_groups.append(null_date_group)
self.logger.info(f"添加无时间戳组: {len(null_date_group)} 个点")
self.logger.info(f"共分为 {len(time_groups)} 个时间组")
return time_groups
def filter_dense_points(self, points_df, grid_size=0.001, distance_threshold=13, time_threshold=timedelta(minutes=5)):
"""
过滤密集点先按时间分组再在每个时间组内过滤
空时间戳的点不进行过滤
Args:
points_df: 点数据
grid_size: 网格大小
distance_threshold: 距离阈值
time_interval: 时间间隔
"""
self.logger.info(f"开始按时间分组过滤密集点 (网格大小: {grid_size}, "
f"距离阈值: {distance_threshold}米, 分组时间间隔: {time_threshold}秒)")
# 按时间分组
time_groups = self._group_by_time(points_df, time_threshold)
# 存储所有要删除的图片
all_to_del_imgs = []
# 对每个时间组进行密集点过滤
for group_idx, group_points in enumerate(time_groups):
# 检查是否为空时间戳组(最后一组)
if group_idx == len(time_groups) - 1 and group_points['date'].isna().any():
self.logger.info(f"跳过无时间戳组 (包含 {len(group_points)} 个点)")
continue
self.logger.info(
f"处理时间组 {group_idx + 1} (包含 {len(group_points)} 个点)")
# 计算该组内的点间距离
sorted_distances = self._get_distances(group_points, grid_size)
group_to_del_imgs = []
# 在每个网格中过滤密集点
for grid, distances in sorted_distances.items():
grid_del_count = 0
while distances:
candidate_img1, candidate_img2, dist = distances[0]
if dist < distance_threshold:
distances.pop(0)
# 获取候选图片的其他最短距离
candidate_img1_dist = None
candidate_img2_dist = None
for distance in distances:
if candidate_img1 in distance:
candidate_img1_dist = distance[2]
break
for distance in distances:
if candidate_img2 in distance:
candidate_img2_dist = distance[2]
break
# 选择要删除的点
if candidate_img1_dist and candidate_img2_dist:
to_del_img = candidate_img1 if candidate_img1_dist < candidate_img2_dist else candidate_img2
group_to_del_imgs.append(to_del_img)
grid_del_count += 1
self.logger.debug(
f"时间组 {group_idx + 1} 网格 {grid} 删除密集点: {to_del_img} (距离: {dist:.2f}米)")
distances = [
d for d in distances if to_del_img not in d]
else:
break
if grid_del_count > 0:
self.logger.info(
f"时间组 {group_idx + 1} 网格 {grid} 删除了 {grid_del_count} 个密集点")
all_to_del_imgs.extend(group_to_del_imgs)
self.logger.info(
f"时间组 {group_idx + 1} 共删除 {len(group_to_del_imgs)} 个密集点")
# 过滤数据
filtered_df = points_df[~points_df['file'].isin(all_to_del_imgs)]
self.logger.info(
f"密集点过滤完成,共删除 {len(all_to_del_imgs)} 个点,剩余 {len(filtered_df)} 个点")
return filtered_df
def filter_isolated_points(self, points_df, threshold_distance=0.001, min_neighbors=6):
"""过滤孤立点"""
self.logger.info(
f"开始过滤孤立点 (距离阈值: {threshold_distance}, 最小邻居数: {min_neighbors})")
coords = points_df[['lat', 'lon']].values
kdtree = KDTree(coords)
neighbors_count = [len(kdtree.query_ball_point(
coord, threshold_distance)) for coord in coords]
isolated_points = []
for i, (_, row) in enumerate(points_df.iterrows()):
if neighbors_count[i] < min_neighbors:
isolated_points.append(row['file'])
self.logger.debug(
f"删除孤立点: {row['file']} (邻居数: {neighbors_count[i]})")
filtered_df = points_df[~points_df['file'].isin(isolated_points)]
self.logger.info(
f"孤立点过滤完成,共删除 {len(isolated_points)} 个点,剩余 {len(filtered_df)} 个点")
return filtered_df

View File

@ -0,0 +1,212 @@
import shutil
import pandas as pd
from shapely.geometry import box
from utils.logger import setup_logger
from utils.gps_extractor import GPSExtractor
import numpy as np
import logging
from datetime import timedelta
import matplotlib.pyplot as plt
import os
import sys
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
class TimeGroupOverlapFilter:
"""基于时间组重叠度的图像过滤器"""
def __init__(self, image_dir: str, output_dir: str, overlap_threshold: float = 0.7):
"""
初始化过滤器
Args:
image_dir: 图像目录
output_dir: 输出目录
overlap_threshold: 重叠阈值默认0.7
"""
self.image_dir = image_dir
self.output_dir = output_dir
self.overlap_threshold = overlap_threshold
self.logger = logging.getLogger('UAV_Preprocess.TimeGroupFilter')
def _group_by_time(self, points_df, time_threshold=timedelta(minutes=5)):
"""按时间间隔对点进行分组"""
if 'date' not in points_df.columns:
self.logger.error("数据中缺少date列")
return []
# 将date为空的行单独作为一组
null_date_group = points_df[points_df['date'].isna()]
valid_date_points = points_df[points_df['date'].notna()]
if not null_date_group.empty:
self.logger.info(f"发现 {len(null_date_group)} 个无时间戳的点,将作为单独分组")
if valid_date_points.empty:
self.logger.warning("没有有效的时间戳数据")
return [null_date_group] if not null_date_group.empty else []
# 按时间排序
valid_date_points = valid_date_points.sort_values('date')
# 计算时间差
time_diffs = valid_date_points['date'].diff()
# 找到时间差超过阈值的位置
time_groups = []
current_group_start = 0
for idx, time_diff in enumerate(time_diffs):
if time_diff and time_diff > time_threshold:
# 添加当前组
current_group = valid_date_points.iloc[current_group_start:idx]
time_groups.append(current_group)
current_group_start = idx
# 添加最后一组
last_group = valid_date_points.iloc[current_group_start:]
if not last_group.empty:
time_groups.append(last_group)
# 如果有空时间戳的点,将其作为最后一组
if not null_date_group.empty:
time_groups.append(null_date_group)
return time_groups
def _get_group_bbox(self, group_df):
"""获取组内点的边界框"""
min_lon = group_df['lon'].min()
max_lon = group_df['lon'].max()
min_lat = group_df['lat'].min()
max_lat = group_df['lat'].max()
return box(min_lon, min_lat, max_lon, max_lat)
def _calculate_overlap(self, box1, box2):
"""计算两个边界框的重叠率"""
if box1.intersects(box2):
intersection_area = box1.intersection(box2).area
smaller_area = min(box1.area, box2.area)
return intersection_area / smaller_area
return 0
def filter_overlapping_groups(self, time_threshold=timedelta(minutes=5)):
"""过滤重叠的时间组"""
# 提取GPS数据
extractor = GPSExtractor(self.image_dir)
gps_points = extractor.extract_all_gps()
# 按时间分组
time_groups = self._group_by_time(gps_points, time_threshold)
# 计算每个组的边界框
group_boxes = []
for idx, group in enumerate(time_groups):
if not group['date'].isna().any(): # 只处理有时间戳的组
bbox = self._get_group_bbox(group)
group_boxes.append((idx, group, bbox))
# 找出需要删除的组
groups_to_delete = set()
for i in range(len(group_boxes)):
if i in groups_to_delete:
continue
idx1, group1, box1 = group_boxes[i]
area1 = box1.area
for j in range(i + 1, len(group_boxes)):
if j in groups_to_delete:
continue
idx2, group2, box2 = group_boxes[j]
area2 = box2.area
overlap_ratio = self._calculate_overlap(box1, box2)
if overlap_ratio > self.overlap_threshold:
# 删除面积较小的组
if area1 < area2:
group_to_delete = idx1
smaller_area = area1
larger_area = area2
else:
group_to_delete = idx2
smaller_area = area2
larger_area = area1
groups_to_delete.add(group_to_delete)
self.logger.info(
f"时间组 {group_to_delete + 1} 与时间组 "
f"{idx2 + 1 if group_to_delete == idx1 else idx1 + 1} "
f"重叠率为 {overlap_ratio:.2f}"
f"面积比为 {smaller_area/larger_area:.2f}"
f"将删除较小面积的组 {group_to_delete + 1}"
)
# 删除重复组的图像
deleted_files = []
for group_idx in groups_to_delete:
group_files = time_groups[group_idx]['file'].tolist()
deleted_files.extend(group_files)
self.logger.info(f"共删除 {len(groups_to_delete)} 个重复时间组,"
f"{len(deleted_files)} 张图像")
# 可视化结果
self._visualize_results(time_groups, groups_to_delete)
return deleted_files
def _visualize_results(self, time_groups, groups_to_delete):
"""可视化过滤结果"""
plt.figure(figsize=(15, 10))
# 生成不同的颜色
colors = plt.cm.rainbow(np.linspace(0, 1, len(time_groups)))
# 绘制所有组的边界框
for idx, (group, color) in enumerate(zip(time_groups, colors)):
if not group['date'].isna().any(): # 只处理有时间戳的组
bbox = self._get_group_bbox(group)
x, y = bbox.exterior.xy
if idx in groups_to_delete:
# 被删除的组用虚线表示
plt.plot(x, y, '--', color=color, alpha=0.6,
label=f'Deleted Group {idx + 1}')
else:
# 保留的组用实线表示
plt.plot(x, y, '-', color=color, alpha=0.6,
label=f'Group {idx + 1}')
# 绘制该组的GPS点
plt.scatter(group['lon'], group['lat'], color=color,
s=30, alpha=0.6)
plt.title("Time Groups and Their Bounding Boxes", fontsize=14)
plt.xlabel("Longitude", fontsize=12)
plt.ylabel("Latitude", fontsize=12)
plt.grid(True)
plt.legend(bbox_to_anchor=(1.05, 1), loc='upper left', fontsize=10)
plt.tight_layout()
# 保存图片
plt.savefig(os.path.join(self.output_dir, 'filter_imgs', 'time_groups_overlap_bbox.png'),
dpi=300, bbox_inches='tight')
plt.close()
if __name__ == '__main__':
# 设置路径
DATASET = r'F:\error_data\20241108134711\3D'
output_dir = r'E:\studio2\ODM_pro\test'
os.makedirs(output_dir, exist_ok=True)
# 设置日志
setup_logger(os.path.dirname(output_dir))
# 创建过滤器并执行过滤
filter = TimeGroupOverlapFilter(DATASET, output_dir, overlap_threshold=0.7)
deleted_files = filter.filter_overlapping_groups(
time_threshold=timedelta(minutes=5))

334
odm_preprocess.py Normal file
View File

@ -0,0 +1,334 @@
import os
import shutil
from datetime import timedelta
from dataclasses import dataclass
from typing import Dict
import matplotlib.pyplot as plt
import pandas as pd
from tqdm import tqdm
from filter.cluster_filter import GPSCluster
from filter.time_group_overlap_filter import TimeGroupOverlapFilter
from filter.gps_filter import GPSFilter
from utils.command_runner import CommandRunner
from utils.gps_extractor import GPSExtractor
from utils.grid_divider import GridDivider
from utils.logger import setup_logger
from utils.visualizer import FilterVisualizer
from post_pro.merge_tif import MergeTif
@dataclass
class PreprocessConfig:
"""预处理配置类"""
image_dir: str
output_dir: str
# 聚类过滤参数
cluster_eps: float = 0.01
cluster_min_samples: int = 5
# 时间组重叠过滤参数
time_group_overlap_threshold: float = 0.7
time_group_interval: timedelta = timedelta(minutes=5)
# 孤立点过滤参数
filter_distance_threshold: float = 0.001 # 经纬度距离
filter_min_neighbors: int = 6
# 密集点过滤参数
filter_grid_size: float = 0.001
filter_dense_distance_threshold: float = 10 # 普通距离,单位:米
filter_time_threshold: timedelta = timedelta(minutes=5)
# 网格划分参数
grid_overlap: float = 0.05
grid_size: float = 500
# 几个pipline过程是否开启
mode: str = "快拼模式"
class ImagePreprocessor:
def __init__(self, config: PreprocessConfig):
self.config = config
# 清理并重建输出目录
if os.path.exists(config.output_dir):
self._clean_output_dir()
self._setup_output_dirs()
# 初始化其他组件
self.logger = setup_logger(config.output_dir)
self.gps_points = None
self.command_runner = CommandRunner(
config.output_dir, mode=config.mode)
self.visualizer = FilterVisualizer(config.output_dir)
def _clean_output_dir(self):
"""清理输出目录"""
try:
shutil.rmtree(self.config.output_dir)
print(f"已清理输出目录: {self.config.output_dir}")
except Exception as e:
print(f"清理输出目录时发生错误: {str(e)}")
raise
def _setup_output_dirs(self):
"""创建必要的输出目录结构"""
try:
# 创建主输出目录
os.makedirs(self.config.output_dir)
# 创建过滤图像保存目录
os.makedirs(os.path.join(self.config.output_dir, 'filter_imgs'))
# 创建日志目录
os.makedirs(os.path.join(self.config.output_dir, 'logs'))
print(f"已创建输出目录结构: {self.config.output_dir}")
except Exception as e:
print(f"创建输出目录时发生错误: {str(e)}")
raise
def extract_gps(self) -> pd.DataFrame:
"""提取GPS数据"""
self.logger.info("开始提取GPS数据")
extractor = GPSExtractor(self.config.image_dir)
self.gps_points = extractor.extract_all_gps()
self.logger.info(f"成功提取 {len(self.gps_points)} 个GPS点")
return self.gps_points
def cluster(self) -> pd.DataFrame:
"""使用DBSCAN对GPS点进行聚类只保留最大的类"""
self.logger.info("开始聚类")
previous_points = self.gps_points.copy()
# 创建聚类器并执行聚类
clusterer = GPSCluster(
self.gps_points, output_dir=self.config.output_dir,
eps=self.config.cluster_eps, min_samples=self.config.cluster_min_samples)
# 获取主要类别的点
self.clustered_points = clusterer.fit()
self.gps_points = clusterer.get_main_cluster(self.clustered_points)
# 获取统计信息并记录
stats = clusterer.get_cluster_stats(self.clustered_points)
self.logger.info(
f"聚类完成:主要类别包含 {stats['main_cluster_points']} 个点,"
f"噪声点 {stats['noise_points']}"
)
# 可视化聚类结果
self.visualizer.visualize_filter_step(
self.gps_points, previous_points, "1-Clustering")
return self.gps_points
def filter_time_group_overlap(self) -> pd.DataFrame:
"""过滤重叠的时间组"""
self.logger.info("开始过滤重叠时间组")
self.logger.info("开始过滤重叠时间组")
previous_points = self.gps_points.copy()
filter = TimeGroupOverlapFilter(
self.config.image_dir,
self.config.output_dir,
overlap_threshold=self.config.time_group_overlap_threshold
)
deleted_files = filter.filter_overlapping_groups(
time_threshold=self.config.time_group_interval
)
# 更新GPS点数据移除被删除的图像
self.gps_points = self.gps_points[~self.gps_points['file'].isin(
deleted_files)]
self.logger.info(f"重叠时间组过滤后剩余 {len(self.gps_points)} 个GPS点")
# 可视化过滤结果
self.visualizer.visualize_filter_step(
self.gps_points, previous_points, "2-Time Group Overlap")
return self.gps_points
# TODO 过滤算法还需要更新
def filter_points(self) -> pd.DataFrame:
"""过滤GPS点"""
self.logger.info("开始过滤GPS点")
filter = GPSFilter(self.config.output_dir)
# 过滤孤立点
previous_points = self.gps_points.copy()
self.logger.info(
f"开始过滤孤立点(距离阈值: {self.config.filter_distance_threshold}, "
f"最小邻居数: {self.config.filter_min_neighbors})"
)
self.gps_points = filter.filter_isolated_points(
self.gps_points,
self.config.filter_distance_threshold,
self.config.filter_min_neighbors,
)
self.logger.info(f"孤立点过滤后剩余 {len(self.gps_points)} 个GPS点")
# 可视化孤立点过滤结果
self.visualizer.visualize_filter_step(
self.gps_points, previous_points, "3-Isolated Points")
# 过滤密集点
previous_points = self.gps_points.copy()
self.logger.info(
f"开始过滤密集点(网格大小: {self.config.filter_grid_size}, "
f"距离阈值: {self.config.filter_dense_distance_threshold})"
)
self.gps_points = filter.filter_dense_points(
self.gps_points,
grid_size=self.config.filter_grid_size,
distance_threshold=self.config.filter_dense_distance_threshold,
time_threshold=self.config.filter_time_threshold,
)
self.logger.info(f"密集点过滤后剩余 {len(self.gps_points)} 个GPS点")
# 可视化密集点过滤结果
self.visualizer.visualize_filter_step(
self.gps_points, previous_points, "4-Dense Points")
return self.gps_points
def divide_grids(self) -> Dict[int, pd.DataFrame]:
"""划分网格"""
self.logger.info(f"开始划分网格 (重叠率: {self.config.grid_overlap})")
self.logger.info(f"开始划分网格 (重叠率: {self.config.grid_overlap})")
grid_divider = GridDivider(overlap=self.config.grid_overlap)
grids = grid_divider.divide_grids(
self.gps_points, grid_size=self.config.grid_size
)
grid_points = grid_divider.assign_to_grids(self.gps_points, grids)
self.logger.info(f"成功划分为 {len(grid_points)} 个网格")
return grid_points
def copy_images(self, grid_points: Dict[int, pd.DataFrame]):
"""复制图像到目标文件夹"""
self.logger.info("开始复制图像文件")
self.logger.info("开始复制图像文件")
for grid_idx, points in grid_points.items():
output_dir = os.path.join(
self.config.output_dir, f"grid_{grid_idx + 1}", "project", "images"
)
os.makedirs(output_dir, exist_ok=True)
for point in tqdm(points, desc=f"复制网格 {grid_idx + 1} 的图像"):
src = os.path.join(self.config.image_dir, point["file"])
dst = os.path.join(output_dir, point["file"])
shutil.copy(src, dst)
self.logger.info(f"网格 {grid_idx + 1} 包含 {len(points)} 张图像")
def merge_tif(self, grid_points: Dict[int, pd.DataFrame]):
"""合并所有网格的TIF影像"""
self.logger.info("开始合并TIF影像")
# 检查是否有多个网格需要合并
if len(grid_points) < 2:
self.logger.info("只有一个网格无需合并TIF影像")
return
input_tif1, input_tif2 = None, None
merge_count = 0
try:
for grid_idx, points in grid_points.items():
grid_tif = os.path.join(
self.config.output_dir,
f"grid_{grid_idx + 1}",
"project",
"odm_orthophoto",
"odm_orthophoto.tif"
)
# 检查TIF文件是否存在
if not os.path.exists(grid_tif):
self.logger.error(
f"网格 {grid_idx + 1} 的TIF文件不存在: {grid_tif}")
continue
if input_tif1 is None:
input_tif1 = grid_tif
self.logger.info(f"设置第一个输入TIF: {input_tif1}")
else:
input_tif2 = grid_tif
output_tif = os.path.join(
self.config.output_dir, "merged_orthophoto.tif")
self.logger.info(
f"开始合并第 {merge_count + 1} 次:\n"
f"输入1: {input_tif1}\n"
f"输入2: {input_tif2}\n"
f"输出: {output_tif}"
)
merge_tif = MergeTif(input_tif1, input_tif2, output_tif)
merge_tif.merge()
merge_count += 1
input_tif1 = output_tif
input_tif2 = None
self.logger.info(
f"TIF影像合并完成共执行 {merge_count} 次合并,"
f"最终输出文件: {input_tif1}"
)
except Exception as e:
self.logger.error(f"TIF影像合并过程中发生错误: {str(e)}", exc_info=True)
raise
def process(self):
"""执行完整的预处理流程"""
try:
self.extract_gps()
self.cluster()
self.filter_time_group_overlap()
self.filter_points()
grid_points = self.divide_grids()
self.copy_images(grid_points)
self.logger.info("预处理任务完成")
self.command_runner.run_grid_commands(
grid_points,
)
# 添加TIF合并步骤
self.merge_tif(grid_points)
except Exception as e:
self.logger.error(f"处理过程中发生错误: {str(e)}", exc_info=True)
raise
if __name__ == "__main__":
# 创建配置
config = PreprocessConfig(
image_dir=r"F:\error_data\20240930091614\project\images",
output_dir=r"G:\20240930091614\output",
cluster_eps=0.01,
cluster_min_samples=5,
# 添加时间组重叠过滤参数
time_group_overlap_threshold=0.7,
time_group_interval=timedelta(minutes=5),
filter_distance_threshold=0.001,
filter_min_neighbors=6,
filter_grid_size=0.001,
filter_dense_distance_threshold=10,
filter_time_threshold=timedelta(minutes=5),
grid_size=1000,
mode="快拼模式",
)
# 创建处理器并执行
processor = ImagePreprocessor(config)
processor.process()

99
post_pro/merge_tif.py Normal file
View File

@ -0,0 +1,99 @@
from osgeo import gdal
import logging
import os
import sys
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
class MergeTif:
def __init__(self, input_tif1, input_tif2, output_tif):
self.input_tif1 = input_tif1
self.input_tif2 = input_tif2
self.output_tif = output_tif
self.logger = logging.getLogger('UAV_Preprocess.MergeTif')
def merge(self):
"""合并两张TIF影像"""
try:
self.logger.info("开始合并TIF影像")
self.logger.info(f"输入影像1: {self.input_tif1}")
self.logger.info(f"输入影像2: {self.input_tif2}")
self.logger.info(f"输出影像: {self.output_tif}")
# 检查输入文件是否存在
if not os.path.exists(self.input_tif1) or not os.path.exists(self.input_tif2):
error_msg = "输入影像文件不存在"
self.logger.error(error_msg)
raise FileNotFoundError(error_msg)
# 打开影像,检查投影是否一致
datasets = [gdal.Open(tif)
for tif in [self.input_tif1, self.input_tif2]]
if None in datasets:
error_msg = "无法打开输入影像文件"
self.logger.error(error_msg)
raise ValueError(error_msg)
projections = [dataset.GetProjection() for dataset in datasets]
self.logger.debug(f"影像1投影: {projections[0]}")
self.logger.debug(f"影像2投影: {projections[1]}")
# 检查投影是否一致
if len(set(projections)) != 1:
error_msg = "影像的投影不一致,请先进行重投影!"
self.logger.error(error_msg)
raise ValueError(error_msg)
# 创建 GDAL Warp 选项
warp_options = gdal.WarpOptions(
format="GTiff",
resampleAlg="average", # 设置重采样方法为平均值
srcNodata=0, # 输入影像中的无效值
dstNodata=0, # 输出影像中的无效值
multithread=True # 启用多线程优化
)
self.logger.info("开始执行影像拼接...")
# 使用 GDAL 的 Warp 方法进行拼接
result = gdal.Warp(
self.output_tif,
[self.input_tif1, self.input_tif2], # 输入多张影像
options=warp_options
)
if result is None:
error_msg = "影像拼接失败"
self.logger.error(error_msg)
raise RuntimeError(error_msg)
# 获取输出影像的基本信息
output_dataset = gdal.Open(self.output_tif)
if output_dataset:
width = output_dataset.RasterXSize
height = output_dataset.RasterYSize
bands = output_dataset.RasterCount
self.logger.info(f"拼接完成,输出影像大小: {width}x{height},波段数: {bands}")
self.logger.info(f"影像拼接成功,输出文件保存至: {self.output_tif}")
except Exception as e:
self.logger.error(f"影像拼接过程中发生错误: {str(e)}", exc_info=True)
raise
if __name__ == "__main__":
from utils.logger import setup_logger
# 定义影像路径
input_tif1 = r"G:\20240930091614\output\grid_1\project\odm_orthophoto\odm_orthophoto.tif"
input_tif2 = r"G:\20240930091614\output\grid_2\project\odm_orthophoto\odm_orthophoto.tif"
output_tif = r"G:\20240930091614\output\merged_orthophoto.tif"
# 设置日志
output_dir = r"E:\studio2\ODM_pro\test"
setup_logger(output_dir)
# 执行拼接
merge_tif = MergeTif(input_tif1, input_tif2, output_tif)
merge_tif.merge()

55
tools/odm_pip_time.py Normal file
View File

@ -0,0 +1,55 @@
from datetime import datetime
import json
def parse_args():
import argparse
parser = argparse.ArgumentParser(description="ODM log time")
parser.add_argument(
"--path", default=r"E:\datasets\UAV\134\project\log.json")
args = parser.parse_args()
return args
def main(args):
# 读取 JSON 文件
with open(args.path, 'r') as file:
data = json.load(file)
# 提取 "stages" 中每个步骤的开始时间和持续时间
stage_timings = []
for i, stage in enumerate(data.get("stages", [])):
stage_name = stage.get("name", "Unnamed Stage")
start_time = stage.get("startTime")
# 获取当前阶段的开始时间
if start_time:
start_dt = datetime.fromisoformat(start_time)
# 获取阶段的结束时间:可以是下一个阶段的开始时间,或当前阶段的 `endTime`(如果存在)
if i + 1 < len(data["stages"]):
end_time = data["stages"][i + 1].get("startTime")
else:
end_time = stage.get("endTime") or data.get("endTime")
if end_time:
end_dt = datetime.fromisoformat(end_time)
duration = (end_dt - start_dt).total_seconds()
stage_timings.append((stage_name, duration))
# 输出每个阶段的持续时间,调整为对齐格式
total_time = 0
print(f"{'Stage Name':<25} {'Duration (seconds)':>15}")
print("=" * 45)
for stage_name, duration in stage_timings:
print(f"{stage_name:<25} {duration:>15.2f}")
total_time += duration
print('Total Time:', total_time)
if __name__ == '__main__':
args = parse_args()
main(args)

51
tools/show_GPS.py Normal file
View File

@ -0,0 +1,51 @@
import os
import sys
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
import matplotlib.pyplot as plt
from preprocess.gps_extractor import GPSExtractor
DATASET = r'F:\error_data\20240930091614\project\images'
if __name__ == '__main__':
extractor = GPSExtractor(DATASET)
gps_points = extractor.extract_all_gps()
# 创建两个子图
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(20, 8))
# 左图:原始散点图
ax1.scatter(gps_points['lon'], gps_points['lat'],
color='blue', marker='o', label='GPS Points')
ax1.set_title("GPS Coordinates of Images", fontsize=14)
ax1.set_xlabel("Longitude", fontsize=12)
ax1.set_ylabel("Latitude", fontsize=12)
ax1.grid(True)
ax1.legend()
# # 右图:按时间排序的轨迹图
# gps_points_sorted = gps_points.sort_values('date')
# # 绘制飞行轨迹线
# ax2.plot(gps_points_sorted['lon'][300:600], gps_points_sorted['lat'][300:600],
# color='blue', linestyle='-', linewidth=1, alpha=0.6)
# # 绘制GPS点
# ax2.scatter(gps_points_sorted['lon'][300:600], gps_points_sorted['lat'][300:600],
# color='red', marker='o', s=30, label='GPS Points')
# 标记起点和终点
# ax2.scatter(gps_points_sorted['lon'].iloc[0], gps_points_sorted['lat'].iloc[0],
# color='green', marker='^', s=100, label='Start')
# ax2.scatter(gps_points_sorted['lon'].iloc[-1], gps_points_sorted['lat'].iloc[-1],
# color='purple', marker='s', s=100, label='End')
ax2.set_title("UAV Flight Trajectory", fontsize=14)
ax2.set_xlabel("Longitude", fontsize=12)
ax2.set_ylabel("Latitude", fontsize=12)
ax2.grid(True)
ax2.legend()
# 调整子图之间的间距
plt.tight_layout()
plt.show()

138
tools/show_GPS_by_time.py Normal file
View File

@ -0,0 +1,138 @@
import os
import sys
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
import matplotlib.pyplot as plt
from datetime import timedelta
import logging
import numpy as np
from preprocess.gps_extractor import GPSExtractor
from preprocess.logger import setup_logger
class GPSTimeVisualizer:
"""按时间组可视化GPS点"""
def __init__(self, image_dir: str, output_dir: str):
self.image_dir = image_dir
self.output_dir = output_dir
self.logger = logging.getLogger('UAV_Preprocess.GPSVisualizer')
def _group_by_time(self, points_df, time_threshold=timedelta(minutes=5)):
"""按时间间隔对点进行分组"""
if 'date' not in points_df.columns:
self.logger.error("数据中缺少date列")
return [points_df]
# 将date为空的行单独作为一组
null_date_group = points_df[points_df['date'].isna()]
valid_date_points = points_df[points_df['date'].notna()]
if not null_date_group.empty:
self.logger.info(f"发现 {len(null_date_group)} 个无时间戳的点,将作为单独分组")
if valid_date_points.empty:
self.logger.warning("没有有效的时间戳数据")
return [null_date_group] if not null_date_group.empty else []
# 按时间排序
valid_date_points = valid_date_points.sort_values('date')
# 计算时间差
time_diffs = valid_date_points['date'].diff()
# 找到时间差超过阈值的位置
time_groups = []
current_group_start = 0
for idx, time_diff in enumerate(time_diffs):
if time_diff and time_diff > time_threshold:
# 添加当前组
current_group = valid_date_points.iloc[current_group_start:idx]
time_groups.append(current_group)
current_group_start = idx
# 添加最后一组
last_group = valid_date_points.iloc[current_group_start:]
if not last_group.empty:
time_groups.append(last_group)
# 如果有空时间戳的点,将其作为最后一组
if not null_date_group.empty:
time_groups.append(null_date_group)
return time_groups
def visualize_time_groups(self, time_threshold=timedelta(minutes=5)):
"""在同一张图上显示所有时间组,用不同颜色区分"""
# 提取GPS数据
extractor = GPSExtractor(self.image_dir)
gps_points = extractor.extract_all_gps()
# 按时间分组
time_groups = self._group_by_time(gps_points, time_threshold)
# 创建图形
plt.figure(figsize=(15, 10))
# 生成不同的颜色
colors = plt.cm.rainbow(np.linspace(0, 1, len(time_groups)))
# 为每个时间组绘制点和轨迹
for idx, (group, color) in enumerate(zip(time_groups, colors)):
if not group['date'].isna().any():
# 有时间戳的组
sorted_group = group.sort_values('date')
# 绘制轨迹线
plt.plot(sorted_group['lon'], sorted_group['lat'],
color=color, linestyle='-', linewidth=1.5, alpha=0.6,
label=f'Flight Path {idx + 1}')
# 绘制GPS点
plt.scatter(sorted_group['lon'], sorted_group['lat'],
color=color, marker='o', s=30, alpha=0.6)
# 标记起点和终点
plt.scatter(sorted_group['lon'].iloc[0], sorted_group['lat'].iloc[0],
color=color, marker='^', s=100,
label=f'Start {idx + 1} ({sorted_group["date"].min().strftime("%H:%M:%S")})')
plt.scatter(sorted_group['lon'].iloc[-1], sorted_group['lat'].iloc[-1],
color=color, marker='s', s=100,
label=f'End {idx + 1} ({sorted_group["date"].max().strftime("%H:%M:%S")})')
else:
# 无时间戳的组
plt.scatter(group['lon'], group['lat'],
color=color, marker='x', s=50, alpha=0.6,
label='No Timestamp Points')
plt.title("GPS Points by Time Groups", fontsize=14)
plt.xlabel("Longitude", fontsize=12)
plt.ylabel("Latitude", fontsize=12)
plt.grid(True)
# 调整图例位置和大小
plt.legend(bbox_to_anchor=(1.05, 1), loc='upper left', fontsize=10)
# 调整布局以适应图例
plt.tight_layout()
# 保存图片
plt.savefig(os.path.join(self.output_dir, 'gps_time_groups_combined.png'),
dpi=300, bbox_inches='tight')
plt.close()
self.logger.info(f"已生成包含 {len(time_groups)} 个时间组的组合可视化图形")
if __name__ == '__main__':
# 设置数据集路径
DATASET = r'F:\error_data\20241108134711\3D'
output_dir = r'E:\studio2\ODM_pro\test'
os.makedirs(output_dir, exist_ok=True)
# 设置日志
setup_logger(os.path.dirname(output_dir))
# 创建可视化器并生成图形
visualizer = GPSTimeVisualizer(DATASET, output_dir)
visualizer.visualize_time_groups(time_threshold=timedelta(minutes=5))

71
utils/command_runner.py Normal file
View File

@ -0,0 +1,71 @@
import os
import logging
import subprocess
import time
from typing import Dict
import pandas as pd
from utils.odm_monitor import ODMProcessMonitor
class CommandRunner:
"""执行网格处理命令的类"""
def __init__(self, output_dir: str, max_retries: int = 3, mode: str = "快拼模式"):
"""
初始化命令执行器
i
Args:
output_dir: 输出目录路径
max_retries: 最大重试次数
"""
self.output_dir = output_dir
self.max_retries = max_retries
self.logger = logging.getLogger('UAV_Preprocess.CommandRunner')
self.monitor = ODMProcessMonitor(max_retries=max_retries, mode=mode)
self.mode = mode
def _run_command(self, grid_idx: int):
"""
执行单个网格的命令
Args:
grid_idx: 网格索引
Raises:
Exception: 当命令执行失败时抛出异常
"""
try:
grid_dir = os.path.join(self.output_dir, f'grid_{grid_idx + 1}')
grid_dir = grid_dir[0].lower() + grid_dir[1:].replace('\\', '/')
if self.mode == "快拼模式":
command = f"docker run -ti --rm -v {grid_dir}:/datasets opendronemap/odm --project-path /datasets project --feature-quality lowest --force-gps --fast-orthophoto --skip-3dmodel"
else:
command = f"docker run -ti --rm -v {grid_dir}:/datasets opendronemap/odm --project-path /datasets project --feature-quality lowest --force-gps"
self.logger.info(f"开始执行命令: {command}")
success, error_msg = self.monitor.run_odm_with_monitor(
command, grid_dir, grid_idx)
if not success:
raise Exception(error_msg)
except Exception as e:
self.logger.error(f"网格 {grid_idx + 1} 处理失败: {str(e)}")
raise
def run_grid_commands(self, grid_points: Dict[int, pd.DataFrame]):
"""
为每个网格顺序运行指定命令
Args:
grid_points: 网格点数据字典键为网格索引值为该网格的点数据
"""
self.logger.info("开始执行网格处理命令")
for grid_idx in grid_points.keys():
try:
self._run_command(grid_idx)
except Exception as e:
self.logger.error(f"网格 {grid_idx + 1} 处理失败,停止后续执行: {str(e)}")
raise

93
utils/gps_extractor.py Normal file
View File

@ -0,0 +1,93 @@
import os
from PIL import Image
import piexif
import logging
import pandas as pd
from datetime import datetime
class GPSExtractor:
"""从图像文件提取GPS坐标和拍摄日期"""
def __init__(self, image_dir):
self.image_dir = image_dir
self.logger = logging.getLogger('UAV_Preprocess.GPSExtractor')
@staticmethod
def _dms_to_decimal(dms):
"""将DMS格式转换为十进制度"""
return dms[0][0] / dms[0][1] + (dms[1][0] / dms[1][1]) / 60 + (dms[2][0] / dms[2][1]) / 3600
@staticmethod
def _parse_datetime(datetime_str):
"""解析EXIF中的日期时间字符串"""
try:
# EXIF日期格式通常为 'YYYY:MM:DD HH:MM:SS'
return datetime.strptime(datetime_str.decode(), '%Y:%m:%d %H:%M:%S')
except Exception:
return None
def get_gps_and_date(self, image_path):
"""提取单张图片的GPS坐标和拍摄日期"""
try:
image = Image.open(image_path)
exif_data = piexif.load(image.info['exif'])
# 提取GPS信息
gps_info = exif_data.get("GPS", {})
lat = lon = None
if gps_info:
lat = self._dms_to_decimal(gps_info.get(2, []))
lon = self._dms_to_decimal(gps_info.get(4, []))
self.logger.debug(f"成功提取图片GPS坐标: {image_path} - 纬度: {lat}, 经度: {lon}")
# 提取拍摄日期
date_info = None
if "Exif" in exif_data:
# 优先使用DateTimeOriginal
date_str = exif_data["Exif"].get(36867) # DateTimeOriginal
if not date_str:
# 备选DateTime
date_str = exif_data["Exif"].get(36868) # DateTimeDigitized
if not date_str:
# 最后使用基本DateTime
date_str = exif_data["0th"].get(306) # DateTime
if date_str:
date_info = self._parse_datetime(date_str)
self.logger.debug(f"成功提取图片拍摄日期: {image_path} - {date_info}")
if not gps_info:
self.logger.warning(f"图片无GPS信息: {image_path}")
if not date_info:
self.logger.warning(f"图片无拍摄日期信息: {image_path}")
return lat, lon, date_info
except Exception as e:
self.logger.error(f"提取图片信息时发生错误: {image_path} - {str(e)}")
return None, None, None
def extract_all_gps(self):
"""提取所有图片的GPS坐标和拍摄日期"""
self.logger.info(f"开始从目录提取GPS坐标和拍摄日期: {self.image_dir}")
gps_data = []
total_images = 0
successful_extractions = 0
for image_file in os.listdir(self.image_dir):
if image_file.lower().endswith('.jpg'):
total_images += 1
image_path = os.path.join(self.image_dir, image_file)
lat, lon, date = self.get_gps_and_date(image_path)
if lat and lon: # 仍然以GPS信息作为主要判断依据
successful_extractions += 1
gps_data.append({
'file': image_file,
'lat': lat,
'lon': lon,
'date': date
})
self.logger.info(f"GPS坐标和拍摄日期提取完成 - 总图片数: {total_images}, 成功提取: {successful_extractions}, 失败: {total_images - successful_extractions}")
return pd.DataFrame(gps_data)

87
utils/grid_divider.py Normal file
View File

@ -0,0 +1,87 @@
import logging
from geopy.distance import geodesic
class GridDivider:
"""划分九宫格,并将图片分配到对应网格"""
def __init__(self, overlap=0.1):
self.overlap = overlap
self.logger = logging.getLogger('UAV_Preprocess.GridDivider')
self.logger.info(f"初始化网格划分器,重叠率: {overlap}")
def divide_grids(self, points_df, grid_size=500):
"""计算边界框并划分九宫格"""
self.logger.info("开始划分九宫格")
min_lat, max_lat = points_df['lat'].min(), points_df['lat'].max()
min_lon, max_lon = points_df['lon'].min(), points_df['lon'].max()
# 计算区域的实际距离(米)
width = geodesic((min_lat, min_lon), (min_lat, max_lon)).meters
height = geodesic((min_lat, min_lon), (max_lat, min_lon)).meters
self.logger.info(
f"区域宽度: {width:.2f}米, 高度: {height:.2f}"
)
# 计算需要划分的网格数量
num_grids_width = int(width / grid_size) if int(width / grid_size) > 0 else 1
num_grids_height = int(height / grid_size) if int(height / grid_size) > 0 else 1
# 计算每个网格对应的经纬度步长
lat_step = (max_lat - min_lat) / num_grids_height
lon_step = (max_lon - min_lon) / num_grids_width
grids = []
for i in range(num_grids_height):
for j in range(num_grids_width):
grid_min_lat = min_lat + i * lat_step - self.overlap * lat_step
grid_max_lat = min_lat + (i + 1) * lat_step + self.overlap * lat_step
grid_min_lon = min_lon + j * lon_step - self.overlap * lon_step
grid_max_lon = min_lon + (j + 1) * lon_step + self.overlap * lon_step
grids.append((grid_min_lat, grid_max_lat, grid_min_lon, grid_max_lon))
self.logger.debug(
f"网格[{i},{j}]: 纬度[{grid_min_lat:.6f}, {grid_max_lat:.6f}], "
f"经度[{grid_min_lon:.6f}, {grid_max_lon:.6f}]"
)
self.logger.info(f"成功划分为 {len(grids)} 个网格 ({num_grids_width}x{num_grids_height})")
return grids
def assign_to_grids(self, points_df, grids):
"""将点分配到对应网格"""
self.logger.info(f"开始将 {len(points_df)} 个点分配到网格中")
grid_points = {i: [] for i in range(len(grids))}
points_assigned = 0
multiple_grid_points = 0
for _, point in points_df.iterrows():
point_assigned = False
for i, (min_lat, max_lat, min_lon, max_lon) in enumerate(grids):
if min_lat <= point['lat'] <= max_lat and min_lon <= point['lon'] <= max_lon:
grid_points[i].append(point.to_dict())
if point_assigned:
multiple_grid_points += 1
else:
points_assigned += 1
point_assigned = True
self.logger.debug(
f"{point['file']} (纬度: {point['lat']:.6f}, 经度: {point['lon']:.6f}) "
f"被分配到网格"
)
# 记录每个网格的点数
for grid_idx, points in grid_points.items():
self.logger.info(f"网格 {grid_idx} 包含 {len(points)} 个点")
self.logger.info(
f"点分配完成: 总点数 {len(points_df)}, "
f"成功分配 {points_assigned} 个点, "
f"{multiple_grid_points} 个点被分配到多个网格"
)
return grid_points

35
utils/logger.py Normal file
View File

@ -0,0 +1,35 @@
import logging
import os
from datetime import datetime
def setup_logger(output_dir):
# 创建logs目录
log_dir = os.path.join(output_dir, 'logs')
# 创建日志文件名(包含时间戳)
timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
log_file = os.path.join(log_dir, f'preprocess_{timestamp}.log')
# 配置日志格式
formatter = logging.Formatter(
'%(asctime)s - %(name)s - %(levelname)s - %(message)s',
datefmt='%Y-%m-%d %H:%M:%S'
)
# 配置文件处理器
file_handler = logging.FileHandler(log_file, encoding='utf-8')
file_handler.setFormatter(formatter)
# 配置控制台处理器
console_handler = logging.StreamHandler()
console_handler.setFormatter(formatter)
# 获取根日志记录器
logger = logging.getLogger('UAV_Preprocess')
logger.setLevel(logging.INFO)
# 添加处理器
logger.addHandler(file_handler)
logger.addHandler(console_handler)
return logger

139
utils/odm_monitor.py Normal file
View File

@ -0,0 +1,139 @@
import os
import time
import psutil
import logging
import subprocess
from typing import Optional, Tuple
class ODMProcessMonitor:
"""ODM进程监控器"""
def __init__(self, max_retries: int = 3, check_interval: int = 10, mode: str = "快拼模式"):
"""
初始化监控器
Args:
max_retries: 最大重试次数
check_interval: 检查间隔
mode: 模式
"""
self.max_retries = max_retries
self.check_interval = check_interval
self.logger = logging.getLogger('UAV_Preprocess.ODMMonitor')
self.mode = mode
def _check_docker_container(self, process_name: str = "opendronemap/odm") -> bool:
"""检查是否有指定的Docker容器在运行"""
try:
result = subprocess.run(
["docker", "ps", "--filter",
f"ancestor={process_name}", "--format", "{{.ID}}"],
capture_output=True,
text=True
)
return bool(result.stdout.strip())
except Exception as e:
self.logger.error(f"检查Docker容器状态时发生错误: {str(e)}")
return False
def _check_success(self, grid_dir: str) -> bool:
"""检查ODM是否执行成功"""
if self.mode == "快拼模式":
success_markers = ['odm_orthophoto', 'odm_georeferencing']
else:
success_markers = ['odm_orthophoto',
'odm_georeferencing', 'odm_texturing']
return all(os.path.exists(os.path.join(grid_dir, 'project', marker)) for marker in success_markers)
def run_odm_with_monitor(self, command: str, grid_dir: str, grid_idx: int) -> Tuple[bool, str]:
"""
运行ODM命令并监控进程
"""
attempt = 0
while attempt < self.max_retries:
try:
self.logger.info(f"网格 {grid_idx + 1}{attempt + 1} 次尝试执行ODM")
# 创建日志文件
log_file = os.path.join(grid_dir, f'odm_attempt_{attempt + 1}.log')
with open(log_file, 'w', encoding='utf-8') as f:
f.write(f"=== ODM处理日志 ===\n开始时间: {time.strftime('%Y-%m-%d %H:%M:%S')}\n\n")
# 启动ODM进程实时获取输出
process = subprocess.Popen(
command,
shell=True,
cwd=grid_dir,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
bufsize=1, # 行缓冲
universal_newlines=True
)
self.logger.info("ODM进程已启动开始监控Docker容器")
# 等待进程启动
time.sleep(10)
# 实时读取输出并写入日志
def log_output(pipe, log_file, prefix=""):
with open(log_file, 'a', encoding='utf-8') as f:
for line in pipe:
f.write(f"{prefix}{line}")
f.flush() # 确保立即写入
# 创建线程读取输出
from threading import Thread
stdout_thread = Thread(target=log_output,
args=(process.stdout, log_file))
stderr_thread = Thread(target=log_output,
args=(process.stderr, log_file, "ERROR: "))
stdout_thread.daemon = True
stderr_thread.daemon = True
stdout_thread.start()
stderr_thread.start()
# 监控Docker容器
while True:
if not self._check_docker_container():
# Docker容器已结束
process.wait() # 等待进程完全结束
# 等待输出线程结束
stdout_thread.join(timeout=5)
stderr_thread.join(timeout=5)
# 记录结束时间
with open(log_file, 'a', encoding='utf-8') as f:
f.write(f"\n=== 处理结束 ===\n结束时间: {time.strftime('%Y-%m-%d %H:%M:%S')}\n")
# 检查是否成功完成
if self._check_success(grid_dir):
self.logger.info(f"网格 {grid_idx + 1} ODM处理成功")
return True, ""
else:
self.logger.warning(
f"网格 {grid_idx + 1}{attempt + 1} 次尝试失败")
break
time.sleep(self.check_interval)
# 如果不是最后一次尝试,等待后重试
if attempt < self.max_retries - 1:
wait_time = (attempt + 1) * 30
self.logger.info(f"等待 {wait_time} 秒后重试...")
time.sleep(wait_time)
attempt += 1
except Exception as e:
error_msg = f"监控进程发生异常: {str(e)}"
self.logger.error(error_msg)
return False, error_msg
error_msg = f"网格 {grid_idx + 1}{self.max_retries} 次尝试后仍然失败,需要人工查看"
self.logger.error(error_msg)
return False, error_msg

121
utils/visualizer.py Normal file
View File

@ -0,0 +1,121 @@
import os
import matplotlib.pyplot as plt
import pandas as pd
import logging
from typing import Optional
class FilterVisualizer:
"""过滤结果可视化器"""
def __init__(self, output_dir: str):
"""
初始化可视化器
Args:
output_dir: 输出目录路径
"""
self.output_dir = output_dir
self.logger = logging.getLogger('UAV_Preprocess.Visualizer')
def visualize_filter_step(self,
current_points: pd.DataFrame,
previous_points: pd.DataFrame,
step_name: str,
save_name: Optional[str] = None):
"""
可视化单个过滤步骤的结果
Args:
current_points: 当前步骤后的点
previous_points: 上一步骤的点
step_name: 步骤名称
save_name: 保存文件名默认为step_name
"""
self.logger.info(f"开始生成{step_name}的可视化结果")
# 找出被过滤掉的点
filtered_files = set(previous_points['file']) - set(current_points['file'])
filtered_points = previous_points[previous_points['file'].isin(filtered_files)]
# 创建图形
plt.figure(figsize=(20, 16))
# 绘制保留的点
plt.scatter(current_points['lon'], current_points['lat'],
color='blue', label='Retained Points',
alpha=0.6, s=50)
# 绘制被过滤的点
if not filtered_points.empty:
plt.scatter(filtered_points['lon'], filtered_points['lat'],
color='red', marker='x', label='Filtered Points',
alpha=0.6, s=100)
# 设置图形属性
plt.title(f"GPS Points After {step_name}\n"
f"(Filtered: {len(filtered_points)}, Retained: {len(current_points)})",
fontsize=14)
plt.xlabel("Longitude", fontsize=12)
plt.ylabel("Latitude", fontsize=12)
plt.grid(True)
# 添加统计信息
stats_text = (
f"Original Points: {len(previous_points)}\n"
f"Filtered Points: {len(filtered_points)}\n"
f"Remaining Points: {len(current_points)}\n"
f"Filter Rate: {len(filtered_points)/len(previous_points)*100:.1f}%"
)
plt.figtext(0.02, 0.02, stats_text, fontsize=10,
bbox=dict(facecolor='white', alpha=0.8))
# 添加图例
plt.legend(loc='upper right', fontsize=10)
# 调整布局
plt.tight_layout()
# 保存图形
save_name = save_name or step_name.lower().replace(' ', '_')
save_path = os.path.join(self.output_dir, 'filter_imgs', f'filter_{save_name}.png')
plt.savefig(save_path, dpi=300, bbox_inches='tight')
plt.close()
self.logger.info(
f"{step_name}过滤可视化结果已保存至 {save_path}\n"
f"过滤掉 {len(filtered_points)} 个点,"
f"保留 {len(current_points)} 个点,"
f"过滤率 {len(filtered_points)/len(previous_points)*100:.1f}%"
)
if __name__ == '__main__':
# 测试代码
import numpy as np
from datetime import datetime
# 创建测试数据
np.random.seed(42)
n_points = 1000
# 生成随机点
test_data = pd.DataFrame({
'lon': np.random.uniform(120, 121, n_points),
'lat': np.random.uniform(30, 31, n_points),
'file': [f'img_{i}.jpg' for i in range(n_points)],
'date': [datetime.now() for _ in range(n_points)]
})
# 随机选择点作为过滤后的结果
filtered_data = test_data.sample(n=800)
# 测试可视化
visualizer = FilterVisualizer('test_output')
os.makedirs('test_output', exist_ok=True)
visualizer.visualize_filter_step(
filtered_data,
test_data,
"Test Filter"
)