System and method for automatically recognizing delivery point information

A method of building machine learning or deep learning data sets for automatically recognizing geographical area information comprising a plurality of geographical area components provided on items, the method comprising:

• first storing, at a first image database, a first plurality of sets of images of geographical area information of distribution items, each of the first plurality of sets of images including an image and OCR information of an entirety of the geographical area information;
• second storing, at a second image database different from the first image database, a second plurality of sets of images of the geographical area information of the distribution items, each of the second plurality of sets of images including images and OCR information of individual geographical area components of the geographical area information, wherein the second plurality of sets of images comprise at least a first subset of images and OCR information of largest geographical area components and a second subset of images and OCR information of intermediate geographical area components;
• converting, at a controller, the first plurality of sets of images and OCR information of the first image database into the second plurality of sets of images and OCR information and storing the second plurality of sets of images and OCR information in the second image database;
• training, using the first subset of images and OCR information of the second image database, a first machine learning or deep learning model to recognize the largest geographical area components;
• and
• training, using the second subset of images and OCR information of the second image database, a second machine learning or deep learning model to recognize the intermediate geographical area components.