Applications of Image Processing Techniques on Palm Leaf Manuscripts – A Survey

Author Name:
Author Email:


In India there are many organizations/institutions committed for protection of ancient palm leaf manuscripts [8] in order to store our precious knowledge writings. As the time passes by, these palm leaves are getting spoiled by artificial and natural elements. With the growing technology worldwide, we are aimed to digitize all these palm leaf manuscripts which are available in our university library. One such objective is to develop an efficient image processing system for effective retrieval of metadata automatically from these manuscripts. Many image processing techniques have been proposed for the efficient data retrieval. Some of the techniques include image enhancement, segmentation, processing, restoration, compression and acquisition. All these techniques have various algorithms which are implemented successfully with desired results. Since one hundred percent correct recognition rate is impossible with noise and other distorted mediums determining the right technique for image processing is very difficult. So, this paper contains information about some image processing techniques and algorithms applied on palm leaf manuscripts proposed by different authors for successful data retrieval.


Manuscript means any information that is hand written. It is generally related to ancient writings of the past men. Palm leaflets [5] are one of the oldest and cheapest materials used for writing. Palm leaf manuscripts (Figure 1) are produced from two main types of palms namely palmyra and talipot. Although palm leaf manuscripts size differs among other parts of the world, the standard size is 48 and 4 centimeters in length and width.

As the time passed by many palm leaves are under severe extinction from various other external elements such as dampness, fungus, ants and cockroaches and many more. SCSVMV University started preserving its Palm Leaf Manuscripts for future generations and made them available in the university library by digitizing then and storing them accordingly. It has a collection of ~5000 manuscripts collected from various sources. The collection mostly comprises of manuscripts written in Grantha Script (~80%) which belong to Tamil, Telugu, Hindi, Kannada, and Malayalam languages.


Fig.1. Palm Leaf Manuscript (Source: Manuscripts Section, SCSVMV University)


Each bundle of leaves is usually tied together with cord threads through two holes pierced through the entire manuscript by the insertion of bamboo strips. The resultant bundle is completed by adding the heavy wooden covers at the either sides of the leaves, also tied by the cords or wrapped with a soft textile cloth. The total number of manuscripts available in SCSVMV University manuscripts section can be visualized in Fig 2.


Although a lot of research works exist in the field of image processing in context with palm leaf manuscripts, but there is a need to prepare an efficient database (DB) for the automation and digitalisation of these manuscripts which can be accessed by all the researchers worldwide (such as a manuscripts DB).  All the techniques discussed above gave good results but more work has to be done on Indian languages due to large character set and linguistic features. From this study, few data sets of manuscripts are used for obtaining the test results but there is a need to train and test huge data sets.


Therefore, it is necessary to prepare an efficient manuscript DB which can automatically generate the metadata from palm leaf manuscripts.



[1] D. Akca, A. Gruen, Re-sequencing a historical palm leaf manuscript with boundary based shape descriptors. Institute of Geodesy and Photogrammetry, Swiss Federal Institute of Technology (ETH) Zuerich, ETH Hoenggerberg, CH 8093 Zuerich, Switzerland.


[2] Olarik Surinta and Rapeeporn Chamchong, Image Segmentation of Historical Handwriting from Palm Leaf Manuscripts. Department of Management Information Systems and Computer Science, Faculty of Informatics, Mahasarakham University, Thailand, 2008.

[3] Panyam Narahari Sastry and Ramakrishnan Krishnan, Classification and Identification of Telugu Handwritten Characters Extracted from Palm Leaves Using Decision Tree Approach. Department of Electronics and Communication Engineering, CBIT, Hyderabad and Indian Institute of Space Science Technology, Trivandram, India, March 2010.

[4] Htwe Pa Pa Win, Phyo Thu Thu Khine, Khin Nwe Ni Tun, Bilingual OCR System for Myanmar and English Scripts with Simultaneous Recognition. International Journal of Scientific & Engineering Research; October 2011, Volume 2, Issue 10, ISSN 2229-5518.

[5] Sunanda Dixit, Dr.H.N.Suresh, South Indian Tamil Language Handwritten Document Text Line Segmentation Technique with aid of Sliding Window and Skewing Operations, Journal of Theoretical and Applied Information Technology; Vol.52 No.2, Dec 2012.

[6] Divakar Yadav, Sonia Sánchez-Cuadrado and Jorge Morato, Optical Character Recognition for Hindi Language using a Neural-network Approach. J Inf Process Syst; March 2013, Vol.9, No.1, pISSN 1976-913X.

[7] Gunjan Singh, Avinash Pokhriyal and Sushma Lehri, Fuzzy Rule Based Classification and Recognition of Handwritten Hindi Curve Script, International journal of Computer Engineering & Technology; 2013, Volume 4, Issue 1, pp. 337 – 357, ISSN Print: 0976 – 6367, ISSN Online: 0976 – 6375.

[8] Rajneesh Rani, Renu Dhir and Gurpreet Singh  Lehal, Modified Gabor Feature Extraction Method       for Word Level Script Identification- Experimentation with Gurumukhi and English Scripts. Dr B R Ambedkar National Institute of Technology Jalandhar and Department of CSE, Punjabi University, Patiala-147002, India, 2013.

[9] Prashanth B U V and P. Narahari Sastry, FPGA based Histogram Equalization Technique to Recognize Characters in Handwritten Scriptures of Palm Leaves. Dept. of ECE, KL University Vijayawada and CBIT, Hyderabad, India, Dec 2013.

[10] R.Vasanth Kumar Mehta, Nagendra Panini Challa, Facilitating enhanced user access through      Palm-leaf Manuscript Digitization – Challenges and solutions. Proceedings of 2nd IEEE International Conference on Electrical, Computer and  Communication Technologies; February 2017, Volume 2, Issue 10, pp. 1034-1039.

[11] Rapeeporn Chamchong, Chun Che Fung, Optimal Selection of Binarization Techniques for the processing of Ancient Palm Leaf Manuscripts, School of Information Technology, Murdoch University 90, South Street, Murdoch, Western Australia 6150, 2010.

[12] Prashanth B U V and P. Narahari Sastry, FPGA based Histogram Equalization Technique to Recognize Characters in Handwritten Scriptures of Palm Leaves. Dept. of ECE, K.L University Vijayawada and CBIT, Hyderabad (Dec 2013). [13] Raju Dara and Urmila Panduga, Telugu Handwritten Isolated Characters Recognition using Two Dimensional Fast Fourier Transform and Support Vector Machine. Department of CSE JNTU, Kakinada, and Senior Software Engineer Cubic Transportation Systems, India (April 2015).

[14] Sk Md Obaidullah, Nibaran Das and Kaushik Roy, Convolution Based Technique for Indic Script Identification from Handwritten Document Images. I.J. Image, Graphics and Signal Processing, (2015), 5, 49-57 April 2015 in MECS.

[15] Rapeeporn Chamchong, Chun Che Fung, Text Line Extraction Using Adaptive Partial Projection for Palm Leaf Manuscripts from Thailand, International Conference on Frontiers in Handwriting Recognition, DOI 10.1109/ICHFR.2012.280, 2012.


26 total views, 0 views today

About the author: tej