HMM based approach for Online Arabic Handwriting recognition

Alshafy, Hozeifa Adam Abd; Mustafa, Mohamed Elhafiz

doi:10.1109/isda.2014.7066288

Cited by 7 publications

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We have used MATLAB for the implementation and evaluation of the technique which is presented in this paper. We use the preprocessing part of the work which present in [11].After that we look two sequences of the handwriting direction (x and y), which are generated by the preprocessing stage, as inputs. Applying the EEDMs and EDMs algorithms, each one as a stand-alone to give us the occurrence of each angle orders.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Extended Feature Extraction Technique (Edge Direction Matrixes) For Online Arabic Handwriting Recognition

Albadawi¹,

Alshafy²

2022

kuej

View full text Add to dashboard Cite

Recognition of Arabic handwriting has attracted the interest of researchers for many years. Until now it has been a challenging research area due to many issues. The feature extraction is an essential stage in the recognition systems of handwriting. The main idea behind this paper is to study EDMs (Edge Direction Matrixes) as a feature extraction technique for Online Arabic Handwriting Recognition. In this study, SUSTOLAH datasets will be used, in which datasets of online Arabic handwriting are presented in Sudan University of Science and Technology. In this paper, satisfactory results have been achieved, where the value of the correlation/regress coefficient for the differences between the variant handwritten characters is found to be -0.01322.

show abstract

Section: Resultsmentioning

confidence: 99%

“…Great research effort has been put in the area of Arabic handwriting recognition [4] [6] [7] [11]. Nevertheless, limited progress has been achieved in this area.…”

Section: Introductionmentioning

confidence: 99%

Extended Feature Extraction Technique (Edge Direction Matrixes) For Online Arabic Handwriting Recognition

Albadawi¹,

Alshafy²

2022

kuej

View full text Add to dashboard Cite

show abstract

“…Recognition performances improved dramatically by introduction of the Deep Learning methods especially in problems where a large amount of training data is available [21], [22], [23]. Yet, HMM based systems are still viable in cases of limited data and computational resources [24] and particular scripts like Arabic [25], [26]. Recurrent Neural Networks and their variants are successfully used for tasks regarding sequential data like online handwriting or speech where data is represented as time series [27], [28], [41].…”

Section: Related Work (İli̇şki̇li̇ çAlişmalar)mentioning

confidence: 99%

Recognition of Online Turkish Handwriting using Transfer Learning

BİLGİN TAŞDEMİR

2023

Gazi Üniversitesi Fen Bilimleri Dergisi Part C: Tasarım Ve Teknoloji

View full text Add to dashboard Cite

We present a recognition system for online Turkish handwriting using transfer learning. Training deep networks requires large amounts of data. Since such a sufficiently large collection of Turkish handwriting samples is not available, So we adopt the transfer learning approach and train and optimize a CNN-BLSTM recognition system first using the standard IAM-On dataset of English handwriting. Then, we fine tune it with Turkish handwriting samples from a smaller dataset. Fine tuning increases the character recognition rate of the final system which is evaluated on 2,041 samples of isolated Turkish words from the initial value of 49% to 85%. The results show that transfer learning can be a solution to the data scarcity problem of online Turkish handwriting.

show abstract

“…The UNIPEN dataset [18] is a collection of characters with recorded pen trajectory information including coordinate data with pen-up/down features, which was used to implement a character recognition model using time delay [19] and Convolutional [33] neural networks. An Arabic recognition system [4] applied HMMs using the SUSTOLAH dataset [34].…”

Section: Positional-data Based Systemsmentioning

confidence: 99%

Towards an IMU-based Pen Online Handwriting Recognizer

Wehbi¹,

Hamann²,

Barth³

et al. 2021

Preprint

View full text Add to dashboard Cite

Most online handwriting recognition systems require the use of specific writing surfaces to extract positional data. In this paper we present a online handwriting recognition system for word recognition which is based on inertial measurement units (IMUs) for digitizing text written on paper. This is obtained by means of a sensor-equipped pen that provides acceleration, angular velocity, and magnetic forces streamed via Bluetooth. Our model combines convolutional and bidirectional LSTM networks, and is trained with the Connectionist Temporal Classification loss that allows the interpretation of raw sensor data into words without the need of sequence segmentation. We use a dataset of words collected using multiple sensor-enhanced pens and evaluate our model on distinct test sets of seen and unseen words achieving a character error rate of 17.97% and 17.08%, respectively, without the use of a dictionary or language model.

show abstract

HMM based approach for Online Arabic Handwriting recognition

Cited by 7 publications

References 7 publications

Extended Feature Extraction Technique (Edge Direction Matrixes) For Online Arabic Handwriting Recognition

Extended Feature Extraction Technique (Edge Direction Matrixes) For Online Arabic Handwriting Recognition

Recognition of Online Turkish Handwriting using Transfer Learning

Towards an IMU-based Pen Online Handwriting Recognizer

Contact Info

Product

Resources

About