Visual odometry step by step ( ZED stereo camera )

1. Initial setting:
  1.  Install OpenCV extra modules
  2.  Setting nsight environment (include and lib path)
  3.  Confirm ZED Serial number (/usr/local/zed/tools/ZED Explorer)
  4.  Key in ZED SN to get camera matrix and rectified images 後來改成用ZED SDK得到camera matrix (每次run程式時,回傳值都有一點誤差,可能是環境溫度等等造成)
  5.  Open ZED Camera and get real-time rectified images
Reference:
2. Feature matching / Stereo matching:
  1. Convert ZED source images to grayscale images
  2. Declare an OpenCV matcher (ex: BFMatcher
  3. Search match points between two images (left and right)
  4. Storing keypoints into vectors 
  5. Using RANSAC method to find good stereo matching points (findHomography OpenCV Fundamental matrix function)
  6. Draw and Display good stereo matching points according RANSAC_mask 
Reference:
  1. OpenCV function (triangulatePoints, convertPointsFromHomogeneous)
  2. Install VTK (Install VTK)
  3. Install PCL (Point Cloud Library)
  4. Install eigen (Install eigen)
Reference:
3.b Get object's 3D position by ZED SDK

Reference:

4. Estimate Camera pose using pnp algorithm
  1. Find initial object's 3D position
  2. Find matching point between lefe image (t-1) and left image (t)                                        Reference: Fundamental matrix RANSAC
  3. Calculate camera pose respect to the initial 3D position (Always use left image)
Reference: 

留言

這個網誌中的熱門文章

Tuing PID parameters in QGroundcontrol (2)

Useful PX4 Parameters

Burn linux image to eMMC storage on Banana Pi M3