Week 3

 

This week was one of the more productive weeks as we managed to:

• -          Connect and run the Operating System of a working Raspberry Pi.
• -          Connect the camera and the Pan-Tilt mount to the Pi board.
• -          Use researched code to operate the camera and take pictures with it.
  
• -          Connected the wiring and get the Pan-Tilt to be recognised by the Raspberry Pi.
• -          Produce code to detect and move the Pan-Tilt mechanical arm using the arrow keys.
• -          Get object detection working with the system. The system is able to detect objects that are shown in front of the camera and display what it thinks the object is with the percentage showing how accurate it is in real time. As shown in the picture above, it is highly accurate in detecting objects.
• -          Detect multiple objects at the same time.
  
• -          Produce a live image feed with labels showing each object and what it is.

 

We ran into the following problems:

• -          Due to the limitations of the hardware, having only 1GB RAM instead of the expected 8GB RAM, the program processes the pictures at roughly between 1-2fps and so there is ghosting, when an object is moved, it will still show the previous location of the box detections even when the object is not there. This is seen in the picture above where the phone is not in the same place as the detection box, due to the phone being moved.
  
• -          We were unable to find any suitable Raspberry Pi boards that were available for ordering. We found a Raspbery Pi 400 keyboard that had an in built Raspberry Pi 4 which could possible work for our system so we ordered that.
  
• -          We were having issues with the power supply. Using the laptop as a power supply was not efficient and there were low voltage warnings. The Raspberry Pi would also shut down at random intervals while the camera and the pan-tilt were both connected to the system and being used. Therefore, we ordered an official power supply for the Raspberry Pi 3B in case the Pi 400 was not compatible with our system, so that we could keep using the Raspberry Pi 3B and hopefully fix the low voltage warnings and random shutdowns.

 

Tasks for next week:

• -          Produce the suitable code needing to have the Pan-Tilt automatically move to follow the object that is being detected.
  
• -          Use libraries and code found to apply object detection algorithms for the camera so that it can detect certain objects held in front of it.
• -          Increase the FPS of the image processing so that the detection system works fluidly and the object detection outputs are without ghosting.