Interfacing USB Camera with ROS Noetic

Project Overview

This repository contains a collection of ROS Noetic projects focused on interfacing a USB camera and utilizing various computer vision techniques. The projects include:

1. Interfacing Camera with ROS
2. Face Detection
3. Color Detection
4. Finger Count
5. Face Counting
6. Simulating Differential Robot with Color
7. Simulating Differential Robot with Finger Count

Project Details

1. Interfacing Camera with ROS

This project demonstrates how to interface a USB camera with ROS Noetic. It includes a publisher node that captures video frames from the camera and publishes them as ROS messages.

• Nodes: camera_publisher.py, camera_subscriber.py
• Topic: video_topic

2. Face Detection

Using OpenCV's Haar Cascade Classifier, this project detects faces in the video feed from the USB camera and displays the detected faces with bounding boxes.

• Nodes: face_pub.py, face_sub.py
• Topic: video_topic

3. Color Detection

This project detects specific colors (red, yellow, blue) in the video feed. It demonstrates how to use HSV color space and thresholding for color detection.

• Nodes: pub_color.py, sub_color.py
• Topic: video_topic
• Detected Colors: Red, Yellow, Blue

4. Finger Count

Using the Mediapipe library, this project detects and counts the number of raised fingers in the video feed. It can be extended to recognize different hand gestures.

• Nodes: hand_pub.py, hand_sub.py
• Topic: video_topic

5. Face Counting

Building on the face detection project, this project counts the number of faces detected in the video feed and displays the count on the screen.

• Nodes: face_c_pub.py, face_c_sub.py
• Topic: video_topic

6. Simulating Differential Robot with Color

This project integrates color detection with robot simulation. The robot performs different actions based on the detected color in the video feed.

• Nodes: diff_color_pub.py, diff_color_sub.py
• Topic: video_topic
• Actions:
  • Red: Turn right
  • Yellow: Move forward
  • Blue: Turn left
  • No Color: Stop

7. Simulating Differential Robot with Finger Count

This project uses finger count detection to control a differential robot's movements. The robot responds to the number of raised fingers detected in the video feed.

• Nodes: diff_hand_pub.py, diff_hand_sub.py
• Topic: video_topic
• Actions:
  • 1 Finger: Move forward
  • 2 Fingers: Move backward or stop
  • 3 Fingers: Move left
  • 4 Fingers: Move right
  • No Fingers: Stop

Installation and Usage

Prerequisites

• ROS Noetic
• OpenCV
• Mediapipe
• USB Camera

Installation

Clone this repository:

Install the required dependencies:

sudo apt-get install ros-noetic-cv-bridge ros-noetic-image-transport
pip install opencv-python mediapipe

Running the Projects

Start the ROS Master:

roscore

Run the Publisher Node:

rosrun your_package_name your_publisher_node.py

Run the Subscriber Nodes:

rosrun your_package_name your_subscriber_node.py

note

Don't forget to add the path in CMakeLists.txt:

catkin_install_python(PROGRAMS
  src/camera_subscriber.py
  src/camera_publisher.py
  src/face_pub.py
  src/face_sub.py
  src/pub_color.py
  src/sub_color.py
  src/hand_sub.py
  src/hand_pub.py
  src/face_c_pub.py
  src/face_c_sub.py
  src/diff_color_pub.py
  src/diff_color_sub.py
  DESTINATION ${CATKIN_PACKAGE_BIN_DESTINATION}
)

If you are using a USB camera then add the port name as shown below:

(before make sure that your camera port and the port name should me same)

videoCaptureObject = cv2.VideoCapture('/dev/video2')

or if it is a default camera then edit code like this :

videoCaptureObject = cv2.VideoCapture(0).

Simulating Differential Robot with Camera

Step-by-Step Instructions

1.Start the ROS Master:

roslaunch master_description gazebo.launch

2.In another terminal, run the mapping launch file:

roslaunch master_description mymapping.launch

3.In another terminal, run the move base launch file:

roslaunch master_description move_base.launch

4.In a fourth terminal, run your publisher node:

rosrun script your_publisher_node.py

5.In a fifth terminal, run your subscriber node:

rosrun script your_subscriber_node.py

Replace your_publisher_node.py and your_subscriber_node.py with the actual filenames of your publisher and subscriber scripts.