Five-Layered Neural Fuzzy Closed-Loop Hybrid Control System with Compound Bayesian Decision-Making Process for Classification Cum Identification of Mixed Connective Conjunct Consonants and Numerals

  • Santosh Kumar HengeEmail author
  • B. Rama
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 553)


The OCR generation systems are most sophisticated active field and interesting conversional discovery for digitalization of handwritten and typed imprecise data into machine detectable characters. The fuzzy logic system processes the data with help of primary-based bunch set of knowledge. Fuzzy logic closed-loop system having very good successful rate of frame work for decision-based functions, can derive the fuzzy rules to build the decision-making procedure and detect the letters human to system to human. Artificial neural networks are compatible and the best area to solve the pattern cum text recognition tasks. The innovative combinational based characteristic of neural fuzzy-based closed-loop hybrid system proposing a five-layered approach with technical ideas, solutions solve the critical problems in the field of character, face, symbol recognition procedure, and estimating the density ratio. Recognition of single text, numbers is easy than the recognition of mixed connective conjunct consonants. Because of their variations, various handwritten pen-stroke pulses, tuning the initial and end position of each conjunct consonant, some consonants are connected and mixed with their left-cum right-side placed conjuncts, numerals, and symbols. Many languages such as Arabic, Hindi, Urdu, Telugu, and Tamil represent syllabic, symbol scripted form, and most of words formed with the mixed conjuncts, mixed cum touched consonants, mixed conjuncts with numerals in their representation. This research approach has proposed the five-layered neural fuzzy closed-loop hybrid system with compound Bayesian decision-making process holding good outcome for classification cum identification of mixed connective conjunct consonants with their numerals. The recognition process can start with categorizing total text into two forms; normal conjunct consonants and mixed connective conjunct consonants. The permutation futures of five-layered neural fuzzy closed-loop hybrid system represent inputs as neurons, convert it into fuzzy set inputs, and then apply the fuzzification process with desirable fuzzy knowledge based rules to produce the required output through responses. The compound Bayesian decision-making process is used to perform the operations of probability of sum to unity to reduce the recognition problems in the mixed connective conjunct consonants.


Bayesian decision-making process (BDMP) Mixed connective conjunct consonants (MCCC) Fuzzy logic controller (FLC) Mixed conjunct consonants (MCjC) Fuzzy artificial neural hybrid system (FANHS) OCR(Optical character recognition) 


  1. 1.
    Razali Bin Abu Bakar,: Development of Online Unconstrained Handwritten Character Recognition Using Fuzzy Logic, Universiti Teknologi MARA.Google Scholar
  2. 2.
    P. Phokharatkul, K. Sankhuangaw, S. Somkuarnpanit, S. Phaiboon, and C. Kimpan: Off-Line Hand Written Thai Character Recognition using Ant-Miner Algorithm. World Academy of Science, Engineering and Technology, 8, 2005, Pg 276–281.Google Scholar
  3. 3.
    Gunvantsinh Gohil, Rekha, Mahesh Goyani: Chain code and holistic features based OCR System for printed devanagari script using ANN and SVM, IJAIA, Vol. 3, No. 1, Jan 2012.Google Scholar
  4. 4.
    P.Vanaja Ranjan,: Efficient Zone Based Feature Extraction Algorithm for Hand Written Numeral Recognition of Four Popular South Indian Scripts, Journal of Theoretical and Applied Information Technology. pg 1171–1181.Google Scholar
  5. 5.
    Santosh Kumar Henge, Laxmikanth, Niranjan: Advanced Fuzzy Logic controller based Mirror-Image-Character-Recognition OCR, LAICEEE. 3101/01/32-32. Pg 261–268.Google Scholar
  6. 6.
    J. Bharathi, Dr. P. Chandrasekar Reddy: Segmentation of Telugu Touching Conjunct Consonants using Overlapping Bounding Boxes, IJCSE, Vol. 5 No. 06 Jun 2013.Google Scholar
  7. 7.
    J.Bharathi, Dr.P.Chandrasekar Reddy: Improvement of Telugu OCR by segmentation of Touching Characters, IJRET, Volume: 03 Issue: 10th Oct-2014.Google Scholar
  8. 8.
    Lotfi A. Zadeh.: Is there a need for fuzzy logic? Department of EECS, University of California, Berkeley, CA 94720–1776, United States, 8 February 2008; 25 February 2008.Google Scholar
  9. 9.
    I.Perfilieya: Fuzzy transforms: a challenge to conventional transforms, in: P.W. Hawkes(Ed), AIEP, vol. 147, Elsevier Academic Press, San Diego 2007, pp. 137–196.Google Scholar
  10. 10.
    L.A. Zadeh: A fuzzy-algorithmic approach to the definition of complex or imprecise concepts, International Journal of Man–Machine Studies 8 (1976) 249–291.Google Scholar
  11. 11.
    Fuzzy Logic Toolbox User’s Guide, The MathWorks Inc., 2001.Google Scholar
  12. 12.
    L.A. Zadeh,: Toward a perception-based theory of probabilistic reasoning with imprecise Probabilities, Journal of Statistical Planning and Inference 105 (2002) 233–264.Google Scholar
  13. 13.
    Mr.Danish Nadeem, Miss. Saleha Rizvi,: Character Recognition using Template Matching. Department of Computer Science, Jamia Millia Islamia, New Delhi.Google Scholar
  14. 14.
    Panyam Narahari, Sastry, Ramakrishnan, Bhagavatula Venkata Sanker Ram: Classification and identification of Telugu Handwritten Characters Extraction from Plam Leaves using Decision Tree Approach”, ARPNJEAS, Vol. 5, No. 3, March 2010.Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2017

Authors and Affiliations

  1. 1.Computer Science DepartmentUniversity College, Kakatiya UniversityHanamkondaIndia

Personalised recommendations