/bag_to_depth

bag_to_depth

Primary LanguagePython

Bag to Depth

ref : https://github.com/nihalsoans91/Bag_to_Depth

This code extracts only pure depth from the bag file.

What is needed

  • A good installation of ROS
  • Numpy
  • python
  • CVbridge

How to execute ?

Pre-Run requirements

Python Code

cd src/bag2rgbdepth/scripts/

Arange the bag files in diffrent folders for each bag file as below

  • Bag folder1
    • bagfile1.bag
  • Bag folder2
    • bagfile2.bag
  • grabrgb.py
  • grabdepth.py
  • truedepthandrgb.sh

How to run

step 1: Run ros.sh using sh ros.sh on one terminal

step 2: Open another terminal and run sh truedepthandrgb.sh

step 3: Wait

What after it executes

The folder strcture now will be

  • Bag folder1
    • bagfile.bag
    • depth_images1
      • dframe1.jpg
      • dframe1.npy
      • ...
      • dframen.jpg
      • dframen.npy
    • rgb_images1
      • frame1.jpg
      • frame1.npy
      • ...
      • framen.jpg
      • framen.npy
  • Bag folder2
    • bagfile2.bag
    • depth_images1
      • dframe1.jpg
      • dframe1.npy
      • ...
      • dframen.jpg
      • dframen.npy
    • rgb_images1
      • frame1.jpg
      • frame1.npy
      • ...
      • framen.jpg
      • framen.npy

C++

First you will need to build the code

Pre-Reqs
  • OpenCV 3
  • ROS desktop full
  • C++ Compiler

Open launch file in launch folder

vim Bag_to_Depth/src/bag2rgbdepth/launch/extractbag_to_rgbd.launch

Edit the line where it says

<param name="folder_extract_location" value="./"/>

Change the value to the folder where you need to extract the images. The code will add the rgb and depth files.

Excute the below bash script

cd Bag_to_Depth/
catkin_make
source devel/setup.bash
roslaunch bag2rgbdepth extractbag_to_rgbd.launch

Meanwhile in another terminal run these two commands

roscore &
rosbag play bagfile1.bag

What after it executes

The folder strcture now will be

  • Bag folder1
    • bagfile.bag
    • depth_images1
      • dframe1.jpg
      • ...
      • dframen.jpg
    • rgb_images1
      • frame1.jpg
      • ...
      • framen.jpg