Functions to read rosbag files

README.md

Installation

1. Windows (ros-noetic with pytorch)

   conda create -y -n XOSight-Pattern2 "python=3.8" "mamba>=0.22.1" -c conda-forge
   conda activate XOSight-Pattern2
   mamba install ros-noetic-desktop -c robostack -c robostack-experimental -c conda-forge --override-channels
   mamba install vs2019_win-64
   conda deactivate
   conda activate XOSight-Pattern2
   mamba install rosdep
   rosdep init
   rosdep update
   
   pip install --extra-index-url https://rospypi.github.io/simple/ rosbag rospy cv_bridge sensor_msgs geometry_msgs roslz4 
   pip install opencv-contrib-python
   python -c "import rosbag, rospy, sensor_msgs, cv_bridge"
   
   conda install pytorch torchvision cudatoolkit=11.6 -c pytorch -c conda-forge
   

   • Ref: To install ros-noetic (Link), To install python mappings i.e. rospypi (Link)

rosbag.py

import tqdm
import rosbag # only python2.7 (http://wiki.ros.org/kinetic/Installation/Ubuntu)
import numpy as np
from sensor_msgs import point_cloud2

## Step1
rosbag_filename = 'myfile.bag'
bag             = rosbag.Bag(rosbag_filename)
print (' - filename          : ', bag.filename)
print (' - get_message_count : ', bag.get_message_count())
print (' - start_time        : ', bag.get_start_time())
print (' - end_time          : ', bag.get_end_time())
print (' - size              : ', bag.size)
print (' - mode              : ', bag.mode)
print ('\n\n')

## Step2 - Method1 - Get all topics
topics          = []
print (' - get_message_count : ', bag.get_message_count())
with tqdm.tqdm_notebook(total = bag.get_message_count()) as pbar:
    for idx, (topic, msg, t) in enumerate(bag.read_messages()):
        pbar.update(1)
        topics.append(topic)
topics = np.array(topics)
print (' - topics : ', np.unique(topics))

## Step2 - Method2 - Get all topics
topics = [topic for topic in bag.get_type_and_topic_info().topics]
for topic in sorted(topics):
    print (topic)

## Step3 - Print messages of a topic
# for idx, (topic, msg, t) in enumerate(bag.read_messages()):
for idx, (topic, msg, t) in enumerate(bag.read_messages(topics=['/velodyne_packets'])):
    print (topic, type(msg), t.to_time())

rosbag_compressed_images.py

import tqdm
import rosbag # only python2.7 (http://wiki.ros.org/kinetic/Installation/Ubuntu)
import numpy as np

import cv2
import matplotlib.pyplot as plt
%matplotlib inline

## Step1 - Read .rosbag
rosbag_filename = 'myfile.bag'
bag             = rosbag.Bag(rosbag_filename)

## Step2 - Read Compressed Images Packets
print (' - start_time        : ', bag.get_start_time())
print (' - end_time          : ', bag.get_end_time())

topics = ['image_raw_left/compressed', 'image_raw_right/compressed']
with tqdm.tqdm_notebook(total = bag.get_message_count(topics)) as pbar:
    for idx, (topic, msg, t) in enumerate(bag.read_messages(topics=topics)):
        pbar.update(1)
        print (topic, type(msg), t.to_time())
        
        img = cv2.imdecode(np.fromstring(msg.data, np.uint8), cv2.IMREAD_COLOR)
        plt.imshow(img)
        filename = os.path.join('/home/ubuntu/data/image_raw_left', str(t.to_time()) + ".jpg")
        cv2.imwrite(filename, img)
        if idx > 150:
            break

rosbag_velodyne.py

import tqdm
import rosbag # only python2.7 (http://wiki.ros.org/kinetic/Installation/Ubuntu)
import numpy as np
from sensor_msgs import point_cloud2
from sensor_msgs.msg import CompressedImage

## Step1 - Read .rosbag
rosbag_filename = 'myfile.bag'
bag             = rosbag.Bag(rosbag_filename)

## Step2 - Read Velodyne Packets
for idx, (topic, msg, t) in enumerate(bag.read_messages(topics=['/velodyne_packets'])):
    print (topic, type(msg), t.to_time())
    packet = msg.serialize
    print (' - header : ', packet.im_self.header)
    data = packet.im_self.packets
    print (' - Total Packets : ', len(data))
    print ('data[0] : ', type(data[0]))
    print ('data[0].stamp : ', type(data[0].stamp))
    print ('data[0].data : ', type(data[0].data))
    break