Robotic and Computer Vision Projects with C++

---

This repository contains projects of computer vision tasks implemented with C++ libraries.

Capture_Data_Kinect1_Kinect2 (Registered)_ZED: Using the Kinect 1, Kinect 2 or a Stereolabs ZED camera, you can save undistorted RGB images (.png), depth maps (.png) and point clouds (.pcd). Additionally, save the camera calibration parameters (.txt). The RGB images of the Kinect 2 are registered (RGB-depth) to their depth map. how use: ./capture __*ZED*__ output: folder /build/SavedFiles/ with all capture data.

Capture_Data_Kinect1_Kinect2 (UnRegistered)_ZED: Using the Kinect 1, Kinect 2 or a Stereolabs ZED camera, you can save undistorted RGB images (.png), depth maps (.png) and point clouds (.pcd). Additionally, save the camera calibration parameters (.txt). The RGB images of the Kinect 2 are unregistered (RGB-depth) to their depth map. how use: ./capture __*ZED*__ output: folder /build/SavedFiles/ with all capture data.

Chessboards_Corners_Detection: Detect and plot the corners of a chessboard using OpenCV. how use: ./corners

Compare_Depth_form_Aruco_and_ZED: Comparison between the depth measured using an Aruco marker and the depth measured with the Stereolabs ZED Camera API. The camera is mounted in a robot Pioneer 3AT. The robot can be drive in two modes: manual (with keyboard) and automatic. how use: ./compareDepth VGA or HD720 or HD1080 or HD2K

Compute_3D_Normals_Keypoints_Correspondences: Compute the normals, keypoints and correspondences in point clouds using PCL. how use: ./correspondences3D robot1.pcd <normals or keypoints or correspondences> <td> <img src="/Images/normals3D.png" width="600" height="100" alt="normals3D" /> </td> <td> <img src="/Images/keypoints3D.png" width="600" height="100" alt="keypoints3D" /> </td> <td> <img src="/Images/correspondences3D.png" width="600" height="100" alt="correspondences3D" /> </td>

Compute_Depth_Error_in_RGBD_Sensors: Detect and plot the corners of a chessboard in real-time using OpenCV and a webcam. inputs: device: Kv1 or Kv2 or ZED_WVGA save_viewers: off or on flag: after or before how use: ./depth_error ../Data/ <device> <save_viewers> <flag>

Compute_Depth_RMS_Error: Computes the depth error between the ground truth (Smart Markers) and the depth computed by a ZED camera. inputs: e1: video in (.svo) format, recorded with a ZED camera mm_our: optimized markers map get by Smart Markers arucoConfig_7: Aruco detection configuration file e1_gt_poses: ground truth camera poses e1_zed_poses: camera poses get by ZED API output: output: file /input/output/e1_r3D_data.txt with timestamp and camera poses (3D position and quaternion). how use: ./depth_rmse ../input/e1.svo ../input/mm_our.yml ../input/arucoConfig_7.yml ../input/e1_gt_poses.txt ../input/e1_zed_poses.txt

Compute_Essential_and_Fundamental_Matrix: Compute the Essential and Fundamental matrix using OpenCV. how use: ./essential_fundamental

Compute_Hausdorff_Distance: Compute Hausdorff distance between two point clouds. how use: ./hausdorff ../Data/input_cloud.pcd ../Data/taget_cloud.pcd Convert_SVO(LeftCam)_Video_to_AVI: inputs: svo_video: video in (.svo) format, recorded with a ZED camera avi_video: name for the output (.avi) video how use: ./svo2avi <svo_video> <output_video> example: ./svo2avi ../Data/test.svo ../Data/test.avi

Correct_Sitting_Position: Demonstrates using PCL and ICP algorithm to correct sitting position using point clouds. inputs: initial_cloud: point cloud of a person in correct sitting position target_cloud: point cloud of a person in correct sitting position mode_visualization: 0 bi-color or 1 rgb-color how use: ./correct_position initial_cloud.pcd target_cloud 1

Crop_Images: Crop images using OpenCV. input: folder ../Images with the images to be crop output: folder ../CroopedImages with the crooped images how use: ./crop ../Images/

Detect_Aruco_Marker: Simple Aruco marker detection on a image. how use: ./aruco_image ../Images/test_0.jpg

Detect_Aruco_with_UVC_Cam: Simple Aruco marker detection in real-time using a webcam. how use: ./aruco_webcam

Detect_Aruco_with_ZED: Simple Aruco marker detection in real-time using ZED camera. how use: ./aruco_zed VGA or HD720 or HD1080 or HD2K

Detect_FractalMarker_ZED: Simple Aruco fractal marker detection in real-time using ZED camera. how use: ./fractal VGA or HD720 or HD1080 or HD2K

Detect_MarkerMap_on_Video: Track a camera using Aruco Marker Map detection. inputs: mm.avi: video with Aruco markers mono_params.yml: intrinsic parameters of a monocular camera marker_size: markers size in meters how use: ./mm_track ../Data/mm.avi ../Data/mm_our.yml ../Data/mono_params.yml 0.1295

Extract_3D_Plane_from_ZED_Video: Extract and visualize the most representative 3D plane in a point cloud. how use: ./extract_plane ../Data/test_0.svo

Extract_3D_Planes_from_PCL_Cloud: Segment the largest 3D planar components from a point cloud how use: ./extract_planes outputs: ../Data/downsampled_cloud.pcd: point cloud after filtering ../Data/detected_plane_0.pcd: point cloud representing the planar component

Installation:

• Dependences (mandatory):

  • ZED SDK 3.0
  • CUDA 10.0
  • OpenCV 3.4.1
  • Aruco 3.0.12
  • PCL 1.8
  • ARIA or ARIACODA
  • UcoSLAM
  • Boost
  • Freenect2
  • Libviso2
  • Open-NI
  • Libusb 1.0

• Download any project and open a terminal ctrl+t:

  $ cd path 
  $ mkdir build & cd build 
  $ cmake .. 
  $ make
  

NOTE:

If you find any of these codes helpful, please share my GitHub and STAR ⭐ this repository to help other enthusiasts to find these tools. Remember, the knowledge must be shared. Otherwise, it is useless and lost in time.