ANPR Automatic number-plate recognition with Python and KNearest

ANPR KNearest creates jpeg img's from TTF font, trains and recognizes licence plate after cleaning and separating chars

EXTRACT TTF FONT CHAR TO JPG

create font - single folder.py

Extract jpg img's from TTF font chars 0-9 and A-Z in single folder as font-eu/0.jpg

create font - subfolders.py

Extract jpg img's from TTF font in subfolders named 0-9 and A-Z with font name on every img as font/0/0-arial.jpg etc

EXTRACTED JPG CHAR TRAINING

training single font done and works

• training font in subfolders is not finished, have issues and need some help to finish this part

LICENCE PLATE RECOGNIZE

recognize.py

• reads JPG img
• cleans img (results in temp folder)
• extracts chars (results in plate-contours folder)
• KNearest return values of nearest dists and shows result

Need asistance to finish "training - font in subfolders.py"