Skip to content

Latest commit

 

History

History
29 lines (20 loc) · 1.87 KB

File metadata and controls

29 lines (20 loc) · 1.87 KB

Surround-View-System-using-4-Fisheye-Cameras-in-Raspberry-Pi-3

By using 4 cameras installed in front, back, right and left of the car, we can see the surroundings in our car in just one screen. These 4 cameras are connected to a multi camera adapter module that the functions is to turn on which camera will be active (this process is sequential).

  • Hardware: Raspberry Pi 3, Multi Camera Adapter V2.2, Fisheye Camera OV05646
  • Tools: Python (OpenCV), I2C Protocol (SDA/SCL), SPI (MISO/MOSI/CLK)
  • Step: 1  Real Hardware _ Chess Board
  1. Take Image

  2. Calibrate Fish Eye 9 9

  3. Convert Bird’s Eye View 3 BEV-Remove-Camera_A_1 3 BEV-Remove-Camera_B_1 3 BEV-Remove-Camera_C_1 3 BEV-Remove-Camera_D_1

  4. Stitch Image 1 1 STITCHING-BEV-OR-6 1 BEV-FINAL-SameSize-Camera_A_1

1 STITCHING-BEV-CAR-FINAL-6 1 BEV-FINAL-SameSize-Camera_A_1

  • Result: FPS = 0.5 with resolution each camera is 640x480
  • This project is collaboration with Foreign Student (Pakistan)