已安装ROS Noetic与cv_bridge仍报ModuleNotFoundError，如何解决？

问题背景 #

已安装ROS Noetic发行版，并通过sudo apt-get install ros-noetic-cv-bridge完成cv_bridge包安装，但在PyCharm的Anaconda虚拟环境（hsi-env）中运行提取rosbag图像帧的代码时，出现ModuleNotFoundError: No module named 'cv_bridge.boost.cv_bridge_boost'错误。

运行代码 #

import os
import argparse
import pdb
import cv2
import rosbag
from sensor_msgs.msg import Image
from cv_bridge import CvBridge

bag_file = './bag_files/20230707_152832.bag'
output_dir= './frames/rgb_bag_output'


image_topic = '/device_0/sensor_1/Color_0/image/data'
# image_topic ='sensor_msgs/Image'

bag = rosbag.Bag(bag_file, "r")
bridge = CvBridge()

#gg =bag.read_messages(topics = '/device_0/sensor_1/Color_0/image/data')
# bag.get_message_count()

count = 0
for topic, msg, t in bag.read_messages(topics=image_topic):
    cv_img = bridge.imgmsg_to_cv2(msg, desired_encoding= "rgb8")#"passthrough")
    cv2.imwrite(os.path.join(output_dir, "frame%06i.png" % count), cv_img)
    print("Wrote image %i" % count)
    count += 1

bag.close() 

错误信息 #

Traceback (most recent call last):
  File "/home/es/PycharmProjects/2-Process-RGBD/extract_bag_frame.py", line 32, in <module>
    cv_img = bridge.imgmsg_to_cv2(msg, desired_encoding= "rgb8")#"passthrough")
  File "/home/es/anaconda3/envs/hsi-env/lib/python3.8/site-packages/cv_bridge/core.py", line 163, in imgmsg_to_cv2
    dtype, n_channels = self.encoding_to_dtype_with_channels(img_msg.encoding)
  File "/home/es/anaconda3/envs/hsi-env/lib/python3.8/site-packages/cv_bridge/core.py", line 99, in encoding_to_dtype_with_channels
    return self.cvtype2_to_dtype_with_channels(self.encoding_to_cvtype2(encoding))
  File "/home/es/anaconda3/envs/hsi-env/lib/python3.8/site-packages/cv_bridge/core.py", line 91, in encoding_to_cvtype2
    from cv_bridge.boost.cv_bridge_boost import getCvType
ModuleNotFoundError: No module named 'cv_bridge.boost.cv_bridge_boost'

错误原因 #

通过sudo apt安装的ros-noetic-cv-bridge是针对ROS自带的系统Python环境（Noetic对应Python3.8）编译的，其boost绑定模块仅存在于系统Python路径中。而你使用的Anaconda虚拟环境有独立的Python路径，无法找到系统cv_bridge的boost依赖；如果虚拟环境中通过pip安装了cv_bridge，该版本通常缺少boost编译的绑定模块，导致报错。

解决方法 #

方法一：切换到ROS自带的Python解释器 #

1. 打开PyCharm，进入Settings → Project: [你的项目名] → Python Interpreter
2. 点击右上角齿轮图标，选择Add → System Interpreter
3. 找到ROS Noetic对应的Python路径，一般为/usr/bin/python3.8
4. 选中该解释器并应用，重新运行代码即可

方法二：在conda环境中重新编译cv_bridge #

1. 创建并进入编译工作空间：

   mkdir -p ~/cv_bridge_ws/src && cd ~/cv_bridge_ws/src
   git clone https://github.com/ros-perception/vision_opencv.git
   cd vision_opencv && git checkout noetic
   cd ../..
   

2. 激活你的conda虚拟环境：

   conda activate hsi-env
   

3. 指定当前conda环境的Python路径编译：

   catkin_make -DPYTHON_EXECUTABLE=$(which python) -DPYTHON_INCLUDE_DIR=$(python -c "from distutils.sysconfig import get_python_inc; print(get_python_inc())") -DPYTHON_LIBRARY=$(python -c "import distutils.sysconfig as sysconfig; print(sysconfig.get_config_var('LIBDIR'))")/libpython3.8.so
   

4. 在PyCharm中添加编译后的cv_bridge路径：
   • 进入Settings → Python Interpreter → Show All
   • 选中你的conda环境，点击Show paths for the selected interpreter
   • 添加路径~/cv_bridge_ws/devel/lib/python3.8/dist-packages，保存后重新运行代码

方法三：手动解析图像数据（无需cv_bridge） #

如果只是临时提取图像帧，可以绕过cv_bridge，直接解析sensor_msgs/Image的原始数据：

import os
import argparse
import pdb
import cv2
import rosbag
import numpy as np
from sensor_msgs.msg import Image

bag_file = './bag_files/20230707_152832.bag'
output_dir= './frames/rgb_bag_output'

image_topic = '/device_0/sensor_1/Color_0/image/data'

bag = rosbag.Bag(bag_file, "r")

count = 0
for topic, msg, t in bag.read_messages(topics=image_topic):
    # 根据图像编码调整dtype和通道数，这里假设是rgb8
    dtype = np.uint8
    channels = 3
    # 从data缓冲区转换为numpy数组
    cv_img = np.frombuffer(msg.data, dtype=dtype).reshape(msg.height, msg.width, channels)
    # 如果是rgb编码，cv2.imwrite默认用bgr，需要转换
    cv_img = cv2.cvtColor(cv_img, cv2.COLOR_RGB2BGR)
    cv2.imwrite(os.path.join(output_dir, "frame%06i.png" % count), cv_img)
    print("Wrote image %i" % count)
    count += 1

bag.close() 

注意：需根据实际图像编码修改dtype和channels，比如mono8对应单通道uint8，16UC1对应单通道uint16。

内容的提问来源于stack exchange，提问作者S.EB