90 lines
3.3 KiB
Python
90 lines
3.3 KiB
Python
import os
|
||
import logging
|
||
import subprocess
|
||
from typing import Dict, Tuple
|
||
import pandas as pd
|
||
|
||
|
||
class ODMProcessMonitor:
|
||
"""ODM处理监控器"""
|
||
|
||
def __init__(self, output_dir: str, mode: str = "快拼模式"):
|
||
self.output_dir = output_dir
|
||
self.logger = logging.getLogger('UAV_Preprocess.ODMMonitor')
|
||
self.mode = mode
|
||
|
||
def _check_success(self, grid_dir: str) -> bool:
|
||
"""检查ODM是否执行成功"""
|
||
success_markers = ['odm_orthophoto', 'odm_georeferencing']
|
||
if self.mode != "快拼模式":
|
||
success_markers.append('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, grid_dir: str, grid_id: tuple, fast_mode: bool = True, produce_dem: bool = False) -> Tuple[bool, str]:
|
||
"""运行ODM命令"""
|
||
if produce_dem and fast_mode:
|
||
self.logger.error("快拼模式下无法生成DEM,请调整生产参数")
|
||
return False, "快拼模式下无法生成DEM,请调整生产参数"
|
||
|
||
self.logger.info(f"开始处理网格 ({grid_id[0]},{grid_id[1]})")
|
||
|
||
# 构建Docker命令
|
||
grid_dir = grid_dir[0].lower()+grid_dir[1:].replace('\\', '/')
|
||
docker_command = (
|
||
f"docker run --gpus all -ti --rm "
|
||
f"-v {grid_dir}:/datasets "
|
||
f"opendronemap/odm:gpu "
|
||
f"--project-path /datasets project "
|
||
f"--max-concurrency 15 "
|
||
f"--force-gps "
|
||
f"--feature-quality lowest "
|
||
f"--orthophoto-resolution 10 "
|
||
)
|
||
|
||
if produce_dem:
|
||
docker_command += (
|
||
f"--dsm "
|
||
f"--dtm "
|
||
)
|
||
|
||
if fast_mode:
|
||
docker_command += (
|
||
f"--fast-orthophoto "
|
||
f"--skip-3dmodel "
|
||
)
|
||
|
||
docker_command += "--rerun-all"
|
||
self.logger.info(docker_command)
|
||
result = subprocess.run(
|
||
docker_command, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
|
||
stdout, stderr = result.stdout.decode(
|
||
'utf-8'), result.stderr.decode('utf-8')
|
||
|
||
self.logger.info(f"==========stdout==========: {stdout}")
|
||
self.logger.error(f"==========stderr==========: {stderr}")
|
||
# 检查执行结果
|
||
if self._check_success(grid_dir):
|
||
self.logger.info(f"网格 ({grid_id[0]},{grid_id[1]}) 处理成功")
|
||
return True, ""
|
||
else:
|
||
self.logger.error(f"网格 ({grid_id[0]},{grid_id[1]}) 处理失败")
|
||
return False, f"网格 ({grid_id[0]},{grid_id[1]}) 处理失败"
|
||
|
||
def process_all_grids(self, grid_points: Dict[tuple, pd.DataFrame], produce_dem: bool):
|
||
"""处理所有网格"""
|
||
self.logger.info("开始执行网格处理")
|
||
for grid_id in grid_points.keys():
|
||
grid_dir = os.path.join(
|
||
self.output_dir, f'grid_{grid_id[0]}_{grid_id[1]}'
|
||
)
|
||
|
||
success, error_msg = self.run_odm_with_monitor(
|
||
grid_dir=grid_dir,
|
||
grid_id=grid_id,
|
||
fast_mode=(self.mode == "快拼模式"),
|
||
produce_dem=produce_dem
|
||
)
|
||
|
||
if not success:
|
||
raise Exception(f"网格 ({grid_id[0]},{grid_id[1]}) 处理失败: {error_msg}")
|