Trackoscope Program

Descriptions of Each Program:

Trackoscope.py (MAIN PROGRAM)

This program is to be run on a laptop. It has the full object-tracking capability and has a user-friendly GUI.

Grapher.py

Taking CSV data and converting to meaningful graphs

AutoFocus.py

This program is to be run on a Raspberry Pi. Uses Picamera library and the Arducam Autofocus Replacement on the Raspberry Pi V2 Camera to autofocus and capture the most focused image.

TrackoscopeRPN.py

This program is to be run on a laptop. It has the full object-tracking capability and has a user-friendly GUI. Opposed to the csrt tracker, this uses a faster neural net tracker (DiSiamRPN)

PiFPSTest.py

This program is to be run on a Raspberry Pi with a Raspberry Pi Camera. Optimizes FPS through threading. Max FPS achieved: 160 fps

Tracking_Rasp.py

This program is to be run on a Raspberry Pi. It has limited object-tracking capability and has a user-friendly GUI. Uses a Pi Camera library to try and optimize FPS.

NonGUITracking.py

This program is to be run on a laptop. It is the earliest stage of object-tracking.

Getting Started

This project uses python virtual environments

Open a command prompt and then install the dependencies to get it working.

pip install -r requirements.txt

Materials

• Arduino Uno
• CNC Shield
• 3 NEMA 17 Stepper Motors
• Webcam/USB Camera