SVM-based hierarchical architectures for handwritten Bangla character recognition

Bhowmik, Tanmay; Ghanty, P.; Roy, Anandarup; Parui, Susanta Kumar

doi:10.1007/s10032-009-0084-x

Cited by 90 publications

(40 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The following accuracies can be reported on the previously mentioned test sets, using the selected models as described in the Model Selection section: 96.25% for Bangla numerals, 80.00% for Bangla basic characters and 97.19% for MNIST dataset. The accuracy on same dataset of basic characters is relatively better than the accuracy is found previously on single stage classification with SVM classifiers [5].…”

Section: Resultscontrasting

confidence: 58%

“…The approximation component is used here as a normalization image in the present recognition problem. In our experiment, we have considered Daubechies wavelet lowpass filter with four coefficients [0.4830, 0.8365, 0.2241, −0.1294] [5]. For a raw input image we first calculate as much as possible the smallest rectangle object region of the image, then normalize it to a square image of size 64 × 64 with an interpolation technique.…”

Section: Feature Extraction With Wavelet Transformmentioning

confidence: 99%

“…From this figure, we can see that the first maximum value of the BIC curve always gives the highest accuracy. In this case, the models λ (4,8), λ (6,5) and λ (8,4) are the models with the highest accuracy, on the basis of the first maxima point of the BIC curve. Since λ (4,8) produces the maximal BIC value with respect to other two models (λ (6,5) and λ (8,4)), λ (4,8) has been chosen as the final model for the numeral zero.…”

Section: Model Selectionmentioning

confidence: 99%

See 2 more Smart Citations

Segmental K-Means Learning with Mixture Distribution for HMM Based Handwriting Recognition

Bhowmik

Oosten

Schomaker

2011

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. This paper investigates the performance of hidden Markov models (HMMs) for handwriting recognition. The Segmental K-Means algorithm is used for updating the transition and observation probabilities, instead of the Baum-Welch algorithm. Observation probabilities are modelled as multi-variate Gaussian mixture distributions. A deterministic clustering technique is used to estimate the initial parameters of an HMM. Bayesian information criterion (BIC) is used to select the topology of the model. The wavelet transform is used to extract features from a grey-scale image, and avoids binarization of the image.

show abstract

Section: Resultscontrasting

confidence: 58%

Section: Feature Extraction With Wavelet Transformmentioning

confidence: 99%

Section: Model Selectionmentioning

confidence: 99%

See 1 more Smart Citation

Segmental K-Means Learning with Mixture Distribution for HMM Based Handwriting Recognition

Bhowmik

Oosten

Schomaker

2011

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…5. It is clear that, when compared to the MNIST dataset, Bangla digits are more complicated and there is more style diversity [7]. For instance, the curly tails in Bangla characters makes the definition of a stable bounding box problematic.…”

Section: Handwritten Digit Classificationmentioning

confidence: 99%

Evaluating automatically parallelized versions of the support vector machine

Codreanu

Droge

Williams

et al. 2014

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYThe Support Vector Machine (SVM) is a supervised learning algorithm used for recognizing patterns in data. It is a very popular technique in Machine Learning and has been successfully used in applications such as image classification, protein classification, and handwriting recognition. However, the computational complexity of the kernelized version of the algorithm grows quadratically with the number of training examples. To tackle this high computational complexity we have developed a directive-based approach that converts a gradient-ascent based training algorithm for the CPU to an efficient GPU implementation. We compare our GPU-based SVM training algorithm to the standard LibSVM CPU implementation, a highly-optimized GPU-LIBSVM implementation, as well as to a directive-based OpenACC implementation. The results on different handwritten digit classification datasets demonstrate an important speed-up for the current approach when compared to the CPU and OpenACC versions. Furthermore, our solution is almost as fast and sometimes even faster than the highly optimized CUBLAS-based GPU-LIBSVM implementation, without sacrificing the algorithm's accuracy.

show abstract

“…Handwritten Tamil Character Recognition system using SVM classifiers were proposed by Shanthi et al [9]. Bhowmik et al [10] provides SVM based hierarchical classification schemes for recognition of handwritten Bangla characters. To the best of our knowledge, the use of SVM in offline handwritten Malayalam characters was reported only in our previous paper [11].…”

Section: Classificationmentioning

confidence: 99%

A System for Offline Recognition of Handwritten Characters in Malayalam Script

John¹,

Balakrishnan²,

Pramod³

2013

IJIGSP

View full text Add to dashboard Cite

Abstract-In this paper, we propose a handwritten character recognition system for Malayalam language. The feature extraction phase consists of gradient and curvature calculation and dimensionality reduction using Principal Component Analysis. Directional information from the arc tangent of gradient is used as gradient feature. Strength of gradient in curvature direction is used as the curvature feature. The proposed system uses a combination of gradient and curvature feature in reduced dimension as the feature vector. For classification, discriminative power of Support Vector Machine (SVM) is evaluated. The results reveal that SVM with Radial Basis Function (RBF) kernel yield the best performance with 96.28% and 97.96% of accuracy in two different datasets. This is the highest accuracy ever reported on these datasets.

show abstract

SVM-based hierarchical architectures for handwritten Bangla character recognition

Cited by 90 publications

References 23 publications

Segmental K-Means Learning with Mixture Distribution for HMM Based Handwriting Recognition

Segmental K-Means Learning with Mixture Distribution for HMM Based Handwriting Recognition

Evaluating automatically parallelized versions of the support vector machine

A System for Offline Recognition of Handwritten Characters in Malayalam Script

Contact Info

Product

Resources

About