Document Image Quality Assessment: A Brief Survey

Abstract

To maintain, control and enhance the quality of document images and minimize the negative impact of degradations on various analysis and processing systems, it is critical to understand the types and sources of degradations and develop reliable methods for estimating the levels of degradations. This paper provides a brief survey of research on the topic of document image quality assessment. We first present a detailed analysis of the types and sources of document degradations. We then review techniques for document image degradation modeling. Finally, we discuss objective measures and subjective experiments that are used to characterize document image quality.

Combining Focus Measure Operators to Predict OCR Accuracy in Mobile-Captured Document Images

“…A good introduction to DIQA subcategories is presented in [2]. This study shows that, despite some prior work on OCR accuracy prediction for scanned document images, only a few approaches deal with camera-or mobile-captured images, and even less are considering out-of-focus blur.…”

“…We are then interested in OCR accuracy prediction, which is a particular case of no-reference Document Image Quality Assessment (DIQA), as defined in [2]. Here "noreference" means that only the test image is available.…”

Subjective quality evaluation of compressed digital compound images

“…DIQA has attracted more attention in the research community recently due to the increasing requirements of digitization of historical or other typewritten documents [33]. Many document image databases are published online, such as University of Washington Dataset [34], Tobacco Database [35], DIQA database [36] and hand-printed document database [37].…”

DocCreator: A New Software for Creating Synthetic Ground-Truthed Document Images

“…As a consequence, it is very important to be aware of the OCR recognition rate before deciding between one of these three solutions. The amount of recent publications on this subject ( [63][64][65][66]) reflects the scientific interest in predicting OCRs recognition rate.…”

“…The difference between the real OCR rate and the one computed on the synthetic versions (Table 2 Column 1 and Column 3) is, on average, only overestimated by 0.03. Most of the success of different existing OCR prediction methods ( [63][64][65][66]) are related to the quality and quantity of the needed ground truth. Our prediction method presented here provides comparable results with the ones form the state of the art.…”