Digitizing paper-based documents to cut down costs and reduce the well-known negative environmental impacts of using and wasting too much paper in offices has led to an increased focus on the systematic electronic scanning of documents through, scanners, mobile phone cameras, etc.
Generally, the quality of the captured document-images is far from good due to a series of challenges related to the performance of the visual sensors and, for camera-based captures, difficult external environmental conditions encountered during the sensing (image capturing) process. Such document-images are mostly hard to read, have low contrast and are corrupted by various artifacts such as noise, blur, shadows, spot lights, etc., just to name a few.
To ensure an acceptable quality of the final document-images that can be perfectly digitalized and involved in various high-level applications based on digital documents, the sensing process must be made much more robust than the raw capture result generated by a purely physical visual sensor. Thus, the physical sensors must be virtually augmented by a series of additional pre- and/or post-processing functional blocks, which mostly involve, amongst others, advanced machine learning techniques.
Paper submissions with innovative and robust approaches are invited for submission as they are needed to solve a series of core issues of relevance for this Special Issue:
Visual sensors related issues w.r.t. document capture or digitization:
Modeling and calibration of visual sensors w.r.t. various distortions
Camera calibration concepts to robustly defocus images
Identification, classification and characterization of visual sensor-related sources of document-image deterioration and distortion
Sharpness quality prediction for mobile-captured document images
Variational models for document-image binarization
Fuzzy models for blur estimation on document images
Adaptive binarization of degraded and/or distorted document images
Rectification and mosaicking of camera-captured document images
Sensor systems for low light document capture and binarization with multiple flash images
Quality assessment of the performance of visual sensors for document capture:
Subjective and objective assessment of the quality of document-images w.r.t. distortions such as blur, noise, contrast, shadow, spot light, etc.
Neurocomputing applications in image quality detection
Post-processing of document-images (captured either by scanners or by mobile phone cameras):
Image quality analysis and enhancement:
Document-image degradation models
Restoration of deteriorated document-images
Quality enhancement of distorted (w.r.t. blur, noise, contrast, shadow, spot light, etc.) document images
Datasets creation for the quality assessment of camera-captured images
Perspective rectification of camera-captured document images
Document image classification and character recognition:
Automated and robust classification of document-images under difficult realistic conditions (i.e. deteriorated or distorted images)
Deep learning applications in robust automated image classification
Impact of image distortion on the readability of QR codes
Robust optical character recognition for distorted and/or deteriorated document-images
Prof. Dr. Kyandoghere Kyamakya
Dr. Fadi Al-Machot
Dr. Ahmad Haj Mosa
Dr. Jean Chamberlain Chedjou