Automated Python Powered Turret
summary
Python code ties OpenCV HOG+SVM person detection to two $18 Logitech webcams and an ORB + BFMatcher stereo pipeline to compute disparity and pick the closest target; an Arduino + servo drives a mounted Nerf gun. Demo works live but has false positives, frame‑rate vs accuracy tradeoffs, a narrow 53° camera FOV, shaky mounts that caused crashes, and torque issues requiring bearings and counterweights. Project succeeds as a proof‑of‑concept with reliability limits on commodity hardware.
AI-written summary · about
chapters (9)
- 0:00Start
- 6:44HOG Pedestrian Detection
- 7:35People detection with Histogram of Oriented Gradients
- 12:00Disparity
- 12:30Correspondence
- 14:11Calibrating HOG detection
- 14:52Lack of knowledge about hardware
- 15:35Setting the gun on the body
- 17:35Lessons Learnt
related talks
21mLeveraging Python for Real-Time Image processing
26mSelf Driving cars with Opencv
49mBuilding a self driving RC car
21mBuilding your own dystopic surveillance state with Python
54mSurveillance Detection Scout – Your Lookout on Autopilot
21mUsing Computer Vision to Solve Interesting Problems
46mDriver Assistant Solutions with
15mRobot Hand Driven by Computer Vision and Reinforcement Learning