Document Image Binarization with Fully Convolutional Neural Networks
Open accessThe DOI matches its Crossref record; 17 of the 17 references it cites were found in a registry or an index.
Abstract
Binarization of degraded historical manuscript images is an important pre-processing step for many document processing tasks. We formulate binarization as a pixel classification learning task and apply a novel Fully Convolutional Network (FCN) architecture that operates at multiple image scales, including full resolution. The FCN is trained to optimize a continuous version of the Pseudo F-measure metric and an ensemble of FCNs outperform the competition winners on 4 of 7 DIBCO competitions. This same binarization technique can also be applied to different domains such as Palm Leaf Manuscripts with good performance. We analyze the performance of the proposed model w.r.t. the architectural hyperparameters, size and diversity of training data, and the input features chosen.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
A Threshold Selection Method from Gray-Level Histograms
Otsu, N. (1979) · IEEE Transactions on Systems, Man, and Cybernetics · doi:10.1109/tsmc.1979.4310076 · Cited by 43,398
Fully convolutional networks for semantic segmentation
Long, J., Shelhamer, E., & Darrell, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298965 · Cited by 38,834
A Computational Approach to Edge Detection
Canny, J. (1986) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.1986.4767851 · Cited by 30,004 · cited 1×
“The unary energy terms are computed from the image Laplacian and the pairwise connections are determined by Canny Edge detection [15].”
MethodAn experimental comparison of min-cut/max- flow algorithms for energy minimization in vision
Boykov, Y., & Kolmogorov, V. (2004) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2004.60 · Cited by 4,577
On the difficulty of training recurrent neural networks
Pascanu, R., Mikolov, T., & Bengio, Y. (2012) · arXiv · doi:10.48550/arxiv.1211.5063 · Cited by 4,310
Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs
Chen, L.-C., Papandreou, G., Kokkinos, I., Murphy, K., & Yuille, A. L. (2014) · arXiv · doi:10.48550/arxiv.1412.7062 · Cited by 4,213 · cited 1×
“combined Conditional Random Fields (CRF) with FCNS to improve localization and consistency of predictions [22], [23].”
BackgroundConditional Random Fields as Recurrent Neural Networks
Zheng, S., Jayasumana, S., Romera-Paredes, B., Vineet, V., Su, Z., Du, D., Huang, C., & Torr, P. H. S. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.179 · Cited by 2,386 · cited 1×
“combined Conditional Random Fields (CRF) with FCNS to improve localization and consistency of predictions [22], [23].”
BackgroundAdaptive document image binarization
Sauvola, J., & Pietikäinen, M. (2000) · Pattern Recognition · doi:10.1016/s0031-3203(99)00055-2 · Cited by 2,339 · cited 2×
“Many common binarization approachs compute local or global thresholds based on image statistics [12], [13].”
BackgroundMethodRobust Document Image Binarization Technique for Degraded Document Images
Su, B., Lu, S., & Tan, C. L. (2013) · IEEE Transactions on Image Processing · doi:10.1109/tip.2012.2231089 · Cited by 282
Document binarization with automatic parameter tuning
Howe, N. R. (2013) · International Journal on Document Analysis and Recognition (IJDAR) · doi:10.1007/s10032-012-0192-x · Cited by 203 · cited 6×
“Qualitative comparison of proposed ensemble of FCNs with state-of-the-art Howe Binarizataion [9].”
BackgroundMethodInfluentialICFHR 2012 Competition on Handwritten Document Image Binarization (H-DIBCO 2012)
Pratikakis, I., Gatos, B., & Ntirogiannis, K. (2012) · 2012 International Conference on Frontiers in Handwriting Recognition · doi:10.1109/icfhr.2012.216 · Cited by 193
ICFHR2016 Handwritten Document Image Binarization Contest (H-DIBCO 2016)
Pratikakis, I., Zagoris, K., Barlas, G., & Gatos, B. (2016) · International Conference on Frontiers in Handwriting Recognition · doi:10.1109/icfhr.2016.0118 · Cited by 184 · cited 2×
“This is a well-studied problem, even for highly degraded documents, as evidenced by the popularity of the Document Image Binarization Content (DIBCO) competitions [1]–[7].”
BackgroundMethodPerformance Evaluation Methodology for Historical Document Image Binarization
Ntirogiannis, K., Gatos, B., & Pratikakis, I. (2013) · IEEE Transactions on Image Processing · doi:10.1109/tip.2012.2219550 · Cited by 167 · cited 3×
“Pseudo F-measure (P-FM) was formulated with the intuition that binarization errors on textual images should be penalized based on how they might obscure individual characters [14].”
MethodInfluentialICFHR2014 Competition on Handwritten Document Image Binarization (H-DIBCO 2014)
Ntirogiannis, K., Gatos, B., & Pratikakis, I. (2014) · International Conference on Frontiers in Handwriting Recognition · doi:10.1109/icfhr.2014.141 · Cited by 108
Insights on the Use of Convolutional Neural Networks for Document Image Binarization
Pastor-Pellicer, J., España-Boquera, S., Zamora-Martínez, F., Afzal, M. Z., & Castro-Bleda, M. J. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-19222-2_10 · Cited by 60 · cited 1×
“explored the use of Convolutional Neural Networks (CNN) to classify each pixel given its 19x19 neighborhood of intesity values [21].”
BackgroundF-Measure as the Error Function to Train Neural Networks
Pastor-Pellicer, J., Zamora-Martínez, F., España-Boquera, S., & Castro-Bleda, M. J. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-38679-4_37 · Cited by 50 · cited 1×
“Using uniform weights yields the more common F-measure, which was explored in [24] as the training objective of neural networks.”
BackgroundLearning document image binarization from data
Wu, Y., Natarajan, P., Rawls, S., & AbdAlmageed, W. (2016) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2016.7533063 · Cited by 34
Metadata
- Authors
- Tensmeyer, C., & Martinez, T.
- Venue
- 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR)
- Publisher
- IEEE
- Published
- 2017-11
- Pages
- 99-104
- OpenAlex
- W2744710511
- Cited by
- 169
- References
- 17
- FWCI
- 5.42
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.