Hybrid Target for Camera-Based Document/Object Capture System

Abstract

The use of mobile phone camera technology in systems for capture of documents or 3D objects by a single exposure is becoming increasingly popular. Compared to conventional flat-bed scanner technology, such systems pose special image quality concerns due to the relatively uncontrolled environment within which they operate. In particular, it is not possible to control ambient lighting, and there is more variability with the internal illumination source and the placement of the document or object to be captured. This leads to higher levels of noise and greater nonuniformity across the image plane. In addition, since the underlying camera and internal illumination technologies are inherently low-cost, each unit is designed to operate closer to the threshold of acceptable quality and there is greater unit-to-unit variability. This makes it necessary to inspect each individual product for acceptable image quality as it comes off the manufacturing line. In order to maintain efficiency in the manufacturing process, this inspection must be completed as autonomously as possible.This paper proposes a specially designed hybrid target and an automatic analysis tool for image quality inspection of camera-based document/object capture systems. The tool can be used in both the research laboratory and on the manufacturing line. The criteria and thresholds for final production are chosen based on visual inspection of image data from hundreds of preproduction units. Currently, our solution is used in mass manufacturing, and shows effectiveness in failing units with low image quality, as well as saving manufacturing time.