A Deep Transfer Learning Approach to Document Image Quality Assessment

Document image quality assessment (DIQA) is an important process for various applications such as optical character recognition (OCR) and document restoration. In this paper we propose a no-reference DIQA model based on a deep convolutional neural network (DCNN), where the rich knowledge of natural scene image characterization of a previously-trained DCNN is exploited towards OCR accuracy oriented document image quality assessment. Following a two-stage deep transfer learning procedure, we fine-tune the knowledge base of the DCNN in the first phase and bring in a task-specific segment consisting of three fully connected (FC) layers in the second phase. Based on the fine-tuned knowledge base, the task-specific segment is trained from scratch to facilitate the application of the transferred knowledge on the new task of document quality assessment. Testing results on a benchmark dataset demonstrate that the proposed model achieves state-of-the-art performance.

[1]  David S. Doermann,et al.  Unsupervised feature learning framework for no-reference image quality assessment , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[2]  Hongyu Li,et al.  CG-DIQA: No-Reference Document Image Quality Assessment Based on Character Gradient , 2018, 2018 24th International Conference on Pattern Recognition (ICPR).

[3]  Yong Liu,et al.  No-reference document image quality assessment based on high order image statistics , 2016, 2016 IEEE International Conference on Image Processing (ICIP).

[4]  Xujun Peng,et al.  Document image OCR accuracy prediction via latent Dirichlet allocation , 2015, 2015 13th International Conference on Document Analysis and Recognition (ICDAR).

[5]  Santanu Chaudhury,et al.  Automatic Selection of Parameters for Document Image Enhancement Using Image Quality Assessment , 2016, 2016 12th IAPR Workshop on Document Analysis Systems (DAS).

[6]  David S. Doermann,et al.  Document Image Quality Assessment: A Brief Survey , 2013, 2013 12th International Conference on Document Analysis and Recognition.

[7]  Alireza Alaei,et al.  Document image quality assessment based on improved gradient magnitude similarity deviation , 2015, 2015 13th International Conference on Document Analysis and Recognition (ICDAR).

[8]  Le Kang,et al.  A deep learning approach to document image quality assessment , 2014, 2014 IEEE International Conference on Image Processing (ICIP).

[9]  David S. Doermann,et al.  Learning features for predicting OCR accuracy , 2012, Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012).

[10]  S. S. Kumar,et al.  Optical character recognition: An overview and an insight , 2014, 2014 International Conference on Control, Instrumentation, Communication and Computational Technologies (ICCICCT).

[11]  Sanghoon Lee,et al.  Deep Convolutional Neural Models for Picture Quality Prediction , 2017 .

[12]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[13]  Tiago Carvalho,et al.  Exposing Computer Generated Images by Using Deep Convolutional Neural Networks , 2017, Signal Process. Image Commun..

[14]  Alireza Alaei,et al.  Document Image Quality Assessment Based on Texture Similarity Index , 2016, 2016 12th IAPR Workshop on Document Analysis Systems (DAS).

[15]  Xujun Peng,et al.  Automated image quality assessment for camera-captured OCR , 2011, 2011 18th IEEE International Conference on Image Processing.

[16]  Mohamed Cheriet,et al.  Subjective and objective quality assessment of degraded document images , 2017 .

[17]  Apostolos Antonacopoulos,et al.  Special issue on the analysis of historical documents , 2007, International Journal of Document Analysis and Recognition (IJDAR).

[18]  Qiang Yang,et al.  A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.

[19]  Nima Tajbakhsh,et al.  Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning? , 2016, IEEE Transactions on Medical Imaging.

[20]  Jean-Marc Ogier,et al.  Combining Focus Measure Operators to Predict OCR Accuracy in Mobile-Captured Document Images , 2014, 2014 11th IAPR International Workshop on Document Analysis Systems.

[21]  Xujun Peng,et al.  Document Image Quality Assessment Using Discriminative Sparse Representation , 2016, 2016 12th IAPR Workshop on Document Analysis Systems (DAS).

[22]  David S. Doermann,et al.  A Dataset for Quality Assessment of Camera Captured Document Images , 2013, CBDAR.

[23]  Jean-Marc Ogier,et al.  Metric-based no-reference quality assessment of heterogeneous document images , 2015, Electronic Imaging.

[24]  Satoshi Naoi,et al.  Automatic filter selection using image quality assessment , 2003, Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings..

[25]  Liangrui Peng,et al.  Attention Based RNN Model for Document Image Quality Assessment , 2017, 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR).

[26]  Patrick Kelly,et al.  Quality assessment and restoration of typewritten document images , 1999, International Journal on Document Analysis and Recognition.