Enhancement of historical printed document images by combining Total Variation regularization and Non-local Means filtering

Likforman-Sulem, Laurence; Darbon, Jérôme; Smith, Elisa H. Barney

doi:10.1016/j.imavis.2011.01.001

Cited by 51 publications

(18 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The proposed method is able to recognize most of the characters of the degraded or blurred document. The performance of proposed method is compare with bilateral filtering with binarization method and it has been found the proposed method work well is adverse condition even when background is highly contrast [14], [15]. Along with better image enhancement we are also able to extract the most of data written on such degraded documents in Sanskrit.…”

Section: Results Analysismentioning

confidence: 91%

A Novel Approach of Data Extraction from Indian Degraded Historical Documents using Gamma Variation and Histogram Balancing Method

Maheshwari¹,

Maloo²,

Parihar³

2015

IJERT

View full text Add to dashboard Cite

Palm leaves manuscripts copper plate compositions, wooden compositions were one of the earliest manifestations of composing media and their utilization as writing material in India has been recorded from as early as the fifth century B.C. until as of late as the late nineteenth century. Palm leaf manuscripts identifying with arts, science, structural engineering, arithmetic, cosmology, astrology, and medication going back a few many years are still accessible for reference today because of numerous continuous endeavors for safeguarding of old archives by libraries and colleges not in India yet all around the globe. Such sort of original copies normally last a couple of hundreds of years yet with time the material degrades and the writing becomes illegible to be valuable in any structure. Advanced Digital Image processing can help enhance the images of these manuscripts in order to empower recovery of the written content from these degraded documents. In this paper we proposed a filter and transform based technique for recovery of data written on such historical original manuscripts. The method uses a dynamically selected pivoting background color in a linear transform to enhance the legibility of the foreground text. At that point a blend of two other image processing algorithms gamma variation method and histogram balancing are applied to the transformed image. The algorithms can be mathematically combined into one or two transformations for computational efficiency. The method is tested on a set of different historical document images in different environmental condition and the result of the proposed method is compared with Bilateral filter with Binarization method show significant improvement in readability with show noteworthy change in clarity. This enhanced image is send to a trained OCR engine for extracting Sanskrit data contents written on manuscripts. The method can likewise be utilized to improve digital images of antiquated, historical, degraded wooden, paper documents.

show abstract

Section: Results Analysismentioning

confidence: 91%

A Novel Approach of Data Extraction from Indian Degraded Historical Documents using Gamma Variation and Histogram Balancing Method

Maheshwari¹,

Maloo²,

Parihar³

2015

IJERT

View full text Add to dashboard Cite

show abstract

“…The NLM filter [3] is used as regularization prior with iterative image reconstruction methods. Using NLM for image restoration (denoising and super-resolution) was presented in [4], [11], [12], and in [5], [9] for image deblurring.…”

Section: Related Workmentioning

confidence: 99%

Deblurring of Document Images Based on Sparse Representations Enhanced by Non-local Means

Nayef

Gomez‐Krämer

Ogier

2014

2014 22nd International Conference on Pattern Recognition

View full text Add to dashboard Cite

Blur is one of the most difficult distortions in camera captured documents. It degrades the visual quality of an image, and makes it difficult to read whether by a human or OCR systems. This paper presents a novel non-blind deblurring method that combines the well known effective techniques of sparse representations and non-local image similarity. The presented problem formulation enables the use of standard sparse coding methods for solving sparse coding-based deblurring when enhanced by a non-local means prior. The method has been tested on both synthetic and real document images degraded with a variety of blur kernels. The resulting deblurred images have high quality in terms of both signal-to-noise ratio and OCR accuracy.

show abstract

“…Indeed, documents suffering from bleed-through have their readability greatly decreased. Moreover, it results in a high OCR error rate [6]. For these reasons, research has been done to measure this defect in order to predict OCR error rates [7], and other works are able to restore degraded documents [8], [2], [9].…”

Section: Introductionmentioning

confidence: 99%

Document Recto-verso Registration Using a Dynamic Time Warping Algorithm

Rabeux

Journet

Philippe

2011

2011 International Conference on Document Analysis and Recognition

View full text Add to dashboard Cite

Recto verso registration is an important step allowing detection of missing digitized pages, or location of the bleed-through defect over a page. An efficient way to restore or evaluate the bleed-through of a digitized document consists in analyzing at the same time both the recto side and the verso side. This method requires the two images to be aligned, registered. Without particular knowledge about document, recto verso registration is complex. Indeed, the only information that we can use to register the two is the bleed-through. Recto verso registration is complex because the recto's bleed-through is a highly degraded version of verso's ink pixels. Therefore, in this particular context, usual image comparison methods [1] are not very relevant. Nevertheless, document recto verso registration algorithms has been proposed [2], [3] [4], but these methods have important time computation costs, are noise sensitive and even fail in some cases where bleed-through is too light. The previous techniques are based on a pixel to pixel approach where the bleed-through is considered to be just a set of grey pixels. In this article, we consider the structure of the ink pixels on the verso page. The recto verso registration method presented here is based on the fact that bleed-through has the same structure that the ink on the verso side. The method registers the recto's bleed-through layout and the verso's ink layout, in two main steps, first a de-skewing algorithm is applied to both pages then, horizontal and vertical profiles are extracted and aligned with a dynamic time warping. The time complexity of our method is linear according to the image size. Moreover, experiments detailed at the end show the accuracy of our method.

show abstract

Enhancement of historical printed document images by combining Total Variation regularization and Non-local Means filtering

Cited by 51 publications

References 33 publications

A Novel Approach of Data Extraction from Indian Degraded Historical Documents using Gamma Variation and Histogram Balancing Method

A Novel Approach of Data Extraction from Indian Degraded Historical Documents using Gamma Variation and Histogram Balancing Method

Deblurring of Document Images Based on Sparse Representations Enhanced by Non-local Means

Document Recto-verso Registration Using a Dynamic Time Warping Algorithm

Contact Info

Product

Resources

About