OCR, Recognition and Information Extraction from Images
summary
Uses OpenCV, Tesseract and Python to extract readable text from photos and PDFs, demonstrating three demos: full-image OCR, region-of-interest extraction, and writing recognized text onto an aerial photo. Presents a deployed license-plate recognition pipeline used in Bolivia that works in production but is sensitive to image capture quality and contrast and needs extra filtering and parameter tuning to handle errors like 1 vs l and stray dashes.
AI-written summary · about
chapters (5)
related talks
20mScalable OCR pipelines using Python,Tensorflow & Tesseract
41mOCR in the Wild World of Social Media
30mIntroduction to Computer Vision with OpenCV
26mChallenge Accepted - How to Escape the Quicksand While Engineering a Computer V...
36mCrash Course in Machine Learning for Natural Language Processing
01h 12mBuilding a live face recognition system
01h 10mBankReport Reader - OCR Tool For Bank Reports
26mDetecting and classifying object images using ruby