Categorizer assisted capture of customer documents using a mobile device

Abstract

A computer-implemented system and method for controlling document image capture are provided. The method includes identifying a categorization model for categorizing image frames and configuring an image capture device, based on the identified categorization model. A flow of frames of a same document captured with the configured image capture device is received and an image representation generated for each of a plurality of frames within the flow of frames. With the identified categorization model, for each of the plurality of frames, a confidence for at least one category is assigned to the frame based on the image representation. A category is assigned to the document when a threshold confidence for one of the at least one categories is assigned. An automatic capture of an image of the document is triggered based on the assigned category.

Description

BACKGROUND[0001]The exemplary embodiment relates to the information processing arts, classification arts, and related arts. It finds particular application in connection with systems and methods for the categorization of customer documents using a mobile device.[0002]Since the advent of digital image sensors, categorization of customer documents captured as a digital image has become common. The captured digital document images may be used for a variety of data extraction purposes, for example, when a customer wishes to subscribe to a new contract with a company, such as a wireless cellphone provider. Much of the required information can be filled out using forms stored electronically on a PC or mobile device, such as a smartphone or tablet. However, a number of physical documents may still be needed, so that they can be scanned for verification. Such documents may include ID cards, driving licenses, passports, proof of address (household utility bills, rent receipts, crossed checks...

AI Technical Summary

Problems solved by technology

When processing the data extracted by such documents, distortions such as noise, dimensions, skew, and rotation make automated categorization and further document processing more difficult.

Document image capturing using the camera of a mobile device may increase this difficulty.

For example, given a digital document image with one of the aforementioned distortions, miscategorization or non-categorization may occur, requiring intervention by a human operator to manually categorize the document.

There are many applications which correct these distortions before categorizing, but these applications require distortion correction after the digital image of the document has been captured, increasing document processing time.

Images