WorkMachine Learning
Election Digit Scanner
Nail handwritten digit recognition for 2024 Indonesia vote recaps with HOG and SVM.
SourceCompleted
This project kills data entry errors in the 2024 Indonesian Presidential Election vote recap with cutting-edge pattern recognition.
Uses Histogram of Oriented Gradients (HOG) for feature extraction and K-Nearest Neighbors (KNN) plus Support Vector Machine (SVM) for classification, hitting over 97% accuracy.
Experiments prove HOG + SVM is the champ, delivering top-tier performance across dataset splits.
What it does
- HOG Feature Magic
- Extracts edges and gradients for pinpoint digit recognition.
- SVM & KNN Power
- Drops 97%+ accuracy with killer classification algorithms.
- Performance Breakdown
- Compares extraction vs. no-extraction for clear wins.
Comments0
Nobody has said anything about this project yet. You could be the first.
Sign in withorto join in.