International Journal of Computer Engineering and Applications, Volume XII, Issue I, Jan. 18, ISSN

Size: px

Start display at page:

Download "International Journal of Computer Engineering and Applications, Volume XII, Issue I, Jan. 18, ISSN"

Bonnie Wendy White
5 years ago
Views:

International Journal of Computer Engineering and Applications, Volume XII, Issue I, Jan. 18, www.ijcea.

of MCA, UBDTCE, Davangere. [Corresponding Author] irshad@ubdtce.org, mjadi14@gmail.

1 International Journal of Computer Engineering and Applications, Volume XII, Issue I, Jan. 18, ISSN A SURVEY ON DIGITIZATION OF HISTORICAL DOCUMENT WITH IMAGE ENHANCEMENT TECHNIQUES ABSTRACT Adarsh M J 1, Md. Irshad Hussian B 2 1 Dept. of MCA, JNNCE, Shimoga 2 Dept. of MCA, UBDTCE, Davangere. [Corresponding Author] irshad@ubdtce.org, mjadi14@gmail.com This paper presents a survey on processing historical documents for digitizing the images and image enhancement techniques that can be used for those digitized images. India is known for its tradition, culture and ritual and there lies a huge knowledge in our vedas,puranas, ayurveda, vedic mathematics, astronomy, etc., in abundant in our historical documents. To bring these documents to digital form a study is required. This paper presents the survey of image enhancement techniques for historical documents, challenges involved in digitalization of these documents. In continuation to this, a survey on script identification is also presented. Keywords: Image Processing, Image enhancement techniques, Historical Documents. [1] INTRODUCTION India is country with huge quantity of traditional background. In India, it is approximately estimated that more than 6 million of documents/manuscripts are available and uncountable inscriptions. Adarsh M J And Md. Irshad Hussian B 308

A SURVEY ON DIGITIZATION OF HISTORICAL DOCUMENT WITH IMAGE ENHANCEMENT TECHNIQUES Figure 1: Historical document These documents are available across the country in form of stone carving, metal

2 A SURVEY ON DIGITIZATION OF HISTORICAL DOCUMENT WITH IMAGE ENHANCEMENT TECHNIQUES Figure 1: Historical document These documents are available across the country in form of stone carving, metal inscriptions, palm leaf manuscripts, clothes and paper scripts. Many organizations are working to find, record, preserve and deliver these to connect India's past with its future, its memory with its aspirations. As these documents are degraded in nature, only digitization process alone is not sufficient. Hence, it is very important to develop suitable algorithm to enhance these documents for deciphering. In reference with this crucial issue, the manuscripts must be securely preserved and kept in soft or hard copies. With that, the ancient Jawi manuscripts are guaranteed of their life and quality and could be shared by interest parties and the future generations. A proactive measure has been taken by the National Library of Malaysia (PNM) to preserve the old manuscripts. Preservation process has been done through several levels of action. Among the actions are cleaning, testing the acid content, de acidification, drying, traditional repair, repair using the leaf casting machine, and binding. However, not all manuscripts which have gone through the preservation process are clear and readable. Sometimes, the documents have contrast problems such as the foregrounds are usually having damaged ink with different background color. Hence, to improve the validity of the manuscripts, the interested parties must deploy digital image processing approaches. It is observed that from the literature survey, many authors have presented a lot of work in the enhancement of the degraded documents using various methods. The rest of the paper is organized as follows. In section 2 we will discuss the challenges present in Enhancement of a Historical Document. In section 3 a brief review of the script identification will be discussed. Section 4 presents the survey of Survey on Image Enhancement Techniques for Historical Documents along with the comparison to the existing methods. Paper will be concluded in section 5. [2] CHALLENGES IN ENHANCEMENT OF A HISTORICAL DOCUMENT: Enhancement of a historical document problem can be formulated as follows: given an input low quality historical document image and the output high quality image for specific applications like script identification, script language identification, text line segmentation, document era identification and can also be for preserving the document for future. When the problem of enhancement of a historical document is considered, some of the challenges could be recognizing characters affected by noise in the documents. Noise like skew shadow, blur, bleed through, paper texture etc. affect the Adarsh M J And Md. Irshad Hussian B 309

3 International Journal of Computer Engineering and Applications, Volume XII, Issue I, Jan. 18, ISSN characters contained in a document and make them degraded. Apart from this, other factors like ageing factors are also contribute considerable amount damage for the historical documents. [3] SURVEY ON SCRIPT IDENTIFICATION India is a country which contains very huge amount of historical documents in libraries, various empires and old temples. Till today they have not been explored because lack understandability of the scripts. Because of this situation, script identification is still a challenge task and which is attracting many researchers. On the other side, low quality and the complexity of these documents (background noise, artifacts due to aging, interfering lines), automatic text line segmentation remains an open research field in computer science. The main objective of this section is to present a survey of existing methods developed for script identification of historical documents during the last decade. Text line segmentation is generally seen as a preprocessing step for tasks such as document structure extraction, printed character or handwriting recognition. Text line segmentation in ancient and historical documents, printed or handwritten, strongly differ because of the document structures. They include various disturbing elements such as holes, spots, writing from the verso appearing on the recto, ornamentation, or seals and medium with which documents are generated. Handwritten documents include narrow spaced lines with overlapping and touching components. Characters and words have unusual and varying shapes, depending on the writer, the period and the place concerned. The vocabulary is also large and may include unusual names and words. Full text recognition is in most cases is not an easy task. The authentication of historical documents in the paleographic sense can also make use of document analysis and text line segmentation. Authentication consists in retrieving writer characteristics independently from document content. It generally consists in dating documents, localizing the place where the document was produced, identifying the writer by using characteristics and features extracted from blank spaces, line orientations and fluctuations, word or character shapes [14] [15]. In Shapiro et al.[16], the global orientation (skew angle) of a handwritten page is first searched by applying a Hough transform on the entire image. Once this skew angle is obtained, projections are achieved along this angle. The number of maxima of the profile gives the number of lines. Low maxima are discarded on their value, which is compared to the highest maxima. Lines are delimited by strips, searching for the minima of projection profiles around each maxima. This technique has been tested on a set of 200 pages within a word segmentation task. In the work of Antonacopoulos and Karatzas [17], each minimum of the profile curve is a potential segmentation point. Potential points are then scored according to their distance to adjacent segmentation points. The reference distance is obtained from the histogram of distances between adjacent potential segmentation points. The highest scored segmentation point is used as an anchor to derive the remaining ones. The method is applied to printed records of the second World War which have regularly spaced text lines. The logical structure is used to derive the text regions where the names of interest can be found. For printed and binarized documents, smearing methods such as the Run-Length Smoothing Algorithm (Wong et al. [18]) can be applied. Consecutive black pixels along the horizontal direction are smeared: i.e. the white space between them is filled with black pixels if their distance is within a predefined threshold. The bounding boxes of the connected components in the smeared image enclose text lines. Adarsh M J And Md. Irshad Hussian B 310

4 A SURVEY ON DIGITIZATION OF HISTORICAL DOCUMENT WITH IMAGE ENHANCEMENT TECHNIQUES A variant of this method adapted to gray level images and applied to printed books from the sixteenth century consists in accumulating the image gradient along the horizontal direction (LeBourgeois [19]). This method has been adapted to old printed documents within the Debora project [20]. For this purpose, numerous adjustments in the method concern the tolerance for character alignment and line justification. Text line patterns are found in the work of Shi and Govindaraju [21] by building a fuzzy run length matrix. At each pixel, the fuzzy run-length is the maximal extent of the background along the horizontal direction. Some foreground pixels may be skipped if their number does not exceed a predefined value. This matrix is threshold to make pieces of text lines appear without ascenders and descenders Parameters have to be accurately and dynamically tuned. [4] SURVEY ON IMAGE ENHANCEMENT TECHNIQUES FOR HISTORICAL DOCUMENT H is to ric a l I m a g e E nh a n c e m e n t 1 B in a riz at i o n a nd T hre s h o lding I m a g e E nh a n c e m e n t 2 A h y brid bin a riz at i o n a nd I m a g e E nh a n c e m e n t 2 N o n - T hre s h o lding M e t h o ds Po st Result Figure 2: Providesthe Block Diagram of Image EnhancementSteps for Old Manuscripts Adaptive Histogram Equalization Histogram Equalization is not suitable for consumer electronics because it could create most of problems. Root Mean Separation is a brightness preservation technique. The preservation ranging is from 0 to 100%. The Dynamic Range value is changed at the output and also the output is based on the picture quality. Here different images having to produce different results. Frequency should be low when the uniform histogram distribution. It offers low frequency. Computation complexity is significantly reduced. Finally the DRSHE could utilize in consumer electronics like LCD and Plasma Display Panel (PDP) TV Histogram Equalization Histogram equalization is broadly used in the field of contrast enhancement. [1, 2] Proposed algorithm mainly focuses on the novel extension and also used to utilize histogram equalization. Ultimate goal is present the brightness value. In this paper newly developed one binary preserved histogram Adarsh M J And Md. Irshad Hussian B 311

5 International Journal of Computer Engineering and Applications, Volume XII, Issue I, Jan. 18, ISSN equalization is proposed. Many applications can be made up of the proposed algorithm. Main aim of proposed algorithm is to reduce the complexity. [3] In this paper is referred to as the generalization of Histogram Equalization. Histogram equalization is not delivered a proper result in such applications. These paper is mainly proposes on brightness preservation techniques. Histogram equalization is significantly introducing the brightness of the image. The consumer electronics field can collect at variety of images is involved. Scalability is the most important property and adjusts the image quality. Ultimate goal of this Histogram Equalization is to allow higher level of brightness preservation. Future work of this paper is to lookout the effective implementation with the use of histogram equalization. [4] Histogram equalization is a one of the useful technique, proposed method and also the comparison of some histogram equalization methods and enhances the contrast, preserve the image as brightness. Different Histogram equalization methods can be used in the images. Each picture is having their own ratio. Experimental results show that two methods M and D are given the best results. [5] Propose a new method known as Brightness Preservation technique. This preservation technique can fulfill the requirement of aforementioned problems. To overcome this problem new mean brightness preservation is added. Each input image is carried out by sub histogram. Performance measure is calculated with the use gray scale preserved brightness images. Future work is recommended to introduce the new measurewhich is also used to evaluate the performance Decor relation Stretch [6] Proposes a practical implementation approach of decor relation and linear contrast image enhancement technology in image processing. The main aim is to extend the medical imaging for visual interpretation such as cerebral. [7] Proposes two pre-processing techniques are implemented. Both two methods are mainly used to improve the classification accuracy. Main aim of this method is to improve the interrupted images and also improve the classification results Image Adjust [8] Proposed method is based on extensive experiment. This paper novel extension of aging scheme is extracted and also the automatic age is to be identified. Human age is estimated based on the genes. The face images patches at different intensity level. Future work is recommended to improve the accuracy. [9] Proposes a new image enhancement method with it is based on the Non-sub Sampled Contour lettransform (NSCT). The proposed algorithm enhances the dynamic range of the image. We have proposed a novel algorithm for multi-scale image enhancement based on the NSCT and also the algorithm can be applied to gray-scale and both color images Image Noise [10] Related work of this paper is related to partial differential equation based schemes for image processing may be easily incorporated in our framework. [11]Film-screen mammography has been the most common and effective technique for the disease for breast cancer. Full-Field Digital Mammography (FFDM) is essential to increase the sensitivity of mammography. In our point of view the proposed methods of this paper is to minimize and avoid alcohol, exercise regularly and also take your supplements daily. Then only you avoid the breast cancer. Adarsh M J And Md. Irshad Hussian B 312

6 A SURVEY ON DIGITIZATION OF HISTORICAL DOCUMENT WITH IMAGE ENHANCEMENT TECHNIQUES [12] This paper is mainly focus on canny edge detector and it is the most popular edge detection technique and also it is the one of the successful edge detector. Future work is recommended to investigate the computing those parameter using property of image such as histogram. New step is also used to increase the computational time. Incorporate peer group and neighbor group consideration can be used to improve edge detection performance. [13]Proposes a plentiful algorithm is used to improve the quality of poor illumination image. Simulation result is purely based on the proposed algorithm. 4.6 Entropy-Based Method There are a few useful methods to calculate background and foreground regions. A prominent method to segment the original image into binary image is called Entropy-based method. Generally, this method is initiated by some threshold values in the range of two variables, namely T1 (background) and T2 (foreground). Some researchers proposed different parameter value arrangements for producing binary image which were also designed for document enhancement purposes. The preserved gray level value is pixel at gray level below T2 value and gray level value within T1 value and T2, whereas the gray level values which are greater than T1 value, are directly eliminated. On the other hand, other research groups have discovered different approaches to encounter color documents. Firstly, they transformed the color documents into gray scale before the binarization process, or directly binarized the color documents. Beside that, another approach called Parameter Estimation Algorithm can also be used to automatically detect the best value for parameter setting (PS). This method executed results from a binary transformed image. It could estimate the different PS values based on any particular image analysis technique. 4.7 Comparison of different enhancement methods: This section presents the comparison of different enhancement methods with their advantages and disadvantages. Table 1 shows the comparisons of various Image enhancement methods. Methods Advantages Disadvantages Adaptive Histogram Equalization It contains low contrast and dark regions. Won t work effectively. Histogram Equalization Décor relation Stretch Image Adjust Image Noise It is a most effective technique for grayscale images. It is originated in the world satellite and aerial mapping. It is used to adjust the image intensity at easily. It is used to reduce the noise from an image easily. Table 1: Comparison of image enhancement methods. But the color images it is a difficult task to work. It is a much complicated process then the other described site. Not able to find the original image. While the dispensable image in low light. Adarsh M J And Md. Irshad Hussian B 313

7 International Journal of Computer Engineering and Applications, Volume XII, Issue I, Jan. 18, ISSN [5] CONCLUSION India is a country which contains very huge amount of historical documents in libraries, various empires and old temples. Enhancement of historical documents can be divided into sub problems like text line segmentation, script identification and enhancement problems. This paper presents a survey of text line segmentation and image enhancement methods for historical documents. The main mail objective of this paper is to make familiar with historical document enhancement problem. REFERENCES [1] Gyu-Hee Park, Haw- Hyan Cho, A Contrast Enhancement Method using Dynamic Range Separate Histogram Equalization, IEEE Transactions on Communication, Networking and Broadcasting, Page: , Publication year: [2] Yeong-Taeg Kim, contrast Enhancement using Brightness Preservation Bi-histogram Equalization, IEEE Transaction on Communication, Networking and Broadcasting, Page: 1-8, Publication year: [3] Soong-Der-chen, Ramli.A.R, Contrast Enhancement using Recursive Mean-Separate Histogram Equalization for Scalable Brightness Preservation, IEEE Transaction on Communication, Networking and Broadcasting, Page: , Publication year: [4] Kim s, Min Chung, Recursively Separate and Weighted Histogram Equalization for Brightness Preservatio n and Contrast Enhancement, IEEE Transaction on Communication, Networking and Broadcasting, Page: , Publication year: [5] Kong.N.S.P, Ibrahim.H, Color Image Enhancement using Brightness Preserving Dynamic Histogram Equalization, IEEE Transaction on Communication, Networking and Broadcasting, Page: , Publication year: [6] Venkata Rasa.M, VR.Anitha, B.AbdhulRahim, VSRD International Journal of Electrical, Electronics and Communication Engineering, VSRD-IJEECE,Page: , Publication year:2001. [7] MichaKalfon and Moshe Porat, A New Approach to Texture Recognition using Decor relation was stretching, Journal of olden doffs Minerva center, Publication year: [8] GuodongGuo, Yun Fu, Charles R. Dyer, Thomas S. Huang, Image based Human Age Estimation by Manifold Learning and Locally Adjusted Robust Regression, IEEE Trasaction on Image Processing, Page: , Publication year:2008. [9] Soyel.H, Mcowan.P.W, Automatic Image Enhancement using intrinsic geometric infor mation, IEEE Trasaction on Communication, Networking and Broadcasting, Page: Publication year: [10] Malladi.R, Sethian.J.A, A Unified Approach to Noise Removal, Image Enhancement and Shape Recovery, IEEE Trasaction on Computing and Processing, Page: Publication year:1996. [11] Mcloughlin.K.J, Bones.P.J, Noise Equalization for Detection of Micro-classification Cluster in Direct digital mammogram images, IEEE Transaction on Computing and Processing, Page: , Publication year: [12] Munasighe.R, Davari.A. Peer groups, Neighbor group and Edge detection, 42nd southeastern symposium on computing and processing, Page:47-50, Publication year:2010. [13] Tzu-Cheng Jen, Sheng-Jyh Wang. Bayesian Structure Preserving Image Contrast Enhancement and its Simplification, IEEE Transaction on Communication, Networking and Broadcasting, Page: , Publication year: [14] Plamondon R., Lorette G. (1989) Automatic signature authentication and writer identification: the state of the art, Pattern Recognition, 22(2): Adarsh M J And Md. Irshad Hussian B 314

8 A SURVEY ON DIGITIZATION OF HISTORICAL DOCUMENT WITH IMAGE ENHANCEMENT TECHNIQUES [15] Bar Yosef I., Kedem K, Dinstein I., Beit-Arie M., Engel E (2004), Classification of Hebrew Calligraphic Handwriting Styles: Preliminary results, First Int. Workshop on Document Image Analysis for Libraries, DIAL 04, Palo Alto. [16] Shapiro V., Gluhchev G., Sgurev V. (1993), Handwritten document image segmentation and analysis, Pattern recognition Letters, 14: [17] Antonacopoulos A., Karatzas D (2004), Document Image analysis for World War II personal records, First Int. Workshop on Document Image Analysis for Libraries, DIAL 04, Palo Alto, pp [18] Wong K., R. Casey, F. Wahl (1982), Document analysis systems, IBM Journal of research and development, Vol 26, no 6. [19]. LeBourgeois F. (1997), Robust multifont OCR system from Gray level Images, 4th International Conference on Document Analysis and Recognition, Ulm. [20]. LeBourgeois F., Emptoz H., Trinh E., Duong J. (2001), Networking Digital Document Images, 6 th International Conference on Document Analysis and Recognition, Seattle. [21]. Shi Z., Govindaraju V. (2004) Line Separation for Complex Document Images Using Fuzzy Runlength, Proc. of the Int. Workshop on Document Image Analysis for Libraries, Palo, Alto, CA, January Adarsh M J And Md. Irshad Hussian B 315

PHASE PRESERVING DENOISING AND BINARIZATION OF ANCIENT DOCUMENT IMAGE

PHASE PRESERVING DENOISING AND BINARIZATION OF ANCIENT DOCUMENT IMAGE Available Online at www.ijcsmc.com International Journal of Computer Science and Mobile Computing A Monthly Journal of Computer Science and Information Technology IJCSMC, Vol. 4, Issue. 7, July 2015, pg.16