An analysis of optical character recognition implementation for ancient Batak characters using K-nearest neighbors principle

Romulus, Puja; Maraden, Yan; Purnamasari, Prima Dewi; Ratna, Anak Agung Putri

doi:10.1109/qir.2015.7374893

Cited by 8 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Devanagari script Naive Bayes, RBF-SVM, and decision tree using HOG and DCT features [5] 33 classes and 5484 characters for training; dataset is unavailable Tamil character Fuzzy median filter for noise removal, a neural network including 3 layers [9] Total class number is unknown; dataset is unavailable Batak script K-Nearest Neighbors [10] Total class number is unknown; dataset is unavailable Vattezhuthu character Image Zoning [6] 237 classes and 5000 characters for training; dataset is unavailable Odia numbers LSTM [7] 10 classes and 5166 characters for training; dataset is unavailable Unlike the previous research on ancient character recognition shown in Table 1, the main challenge of our work is that the total number of characters to be classified is 681 characters, and there is only one reference sample for each character. We aim to find a general method that uses public and sufficient data resources from other domains to perform the same retrieval task for public ancient character scripts.…”

Section: Methods Task and Data Availabilitymentioning

confidence: 99%

A Prototypical Network-Based Approach for Low-Resource Font Typeface Feature Extraction and Utilization

Batjargal

Maeda

2021

Data

View full text Add to dashboard Cite

This paper introduces a framework for retrieving low-resource font typeface databases by handwritten input. A new deep learning model structure based on metric learning is proposed to extract the features of a character typeface and predict the category of handwrittten input queries. Rather than using sufficient training data, we aim to utilize ancient character font typefaces with only one sample per category. Our research aims to achieve decent retrieval performances over more than 600 categories of handwritten characters automatically. We consider utilizing generic handcrafted features to train a model to help the voting classifier make the final prediction. The proposed method is implemented on the ‘Shirakawa font oracle bone script’ dataset as an isolated ancient-character-recognition system based on free ordering and connective strokes. We evaluate the proposed model on several standard character and symbol datasets. The experimental results showed that the proposed method provides good performance in extracting the features of symbols or characters’ font images necessary to perform further retrieval tasks. The demo system has been released, and it requires only one sample for each character to predict the user input. The extracted features have a better effect in finding the highest-ranked relevant item in retrieval tasks and can also be utilized in various technical frameworks for ancient character recognition and can be applied to educational application development.

show abstract

Section: Methods Task and Data Availabilitymentioning

confidence: 99%

A Prototypical Network-Based Approach for Low-Resource Font Typeface Feature Extraction and Utilization

Batjargal

Maeda

2021

Data

View full text Add to dashboard Cite

show abstract

“…In this result Ga = ga and has consistency in capital. After the calculation results are obtained, the contrastive loss function will determine the correct result regarding the meaning of the word by looking for the smallest value result [24].…”

Section: One-shot Learningmentioning

confidence: 99%

One-shot learning Batak Toba character recognition using siamese neural network

et al. 2023

View full text Add to dashboard Cite

Siamese neural network (SINN) is an image processing model that compares the scores of two patterns. The SINN algorithm is a combination of the use of the double convolutional neural network (CNN) algorithm. By combined SINN with a one-shot learning algorithm, we can build an image model without requiring thousands of images for training. The test results from the SINN algorithm and one-shot learning show that this process was successful in matching the two data but was unable to produce labels from the data being tested. Because of this, the researcher decided to continue the implementation process using the CNN algorithm combined with single shot detection (SSD). By using a dataset of 5000, the recognition and translation of the Toba Batak script was successful. The percentage of average accuracy results from CNN and SSD in recognizing Toba Batak characters is 84.08% for single characters and 74.13% for mixed characters. While the percentage of average accuracy results for testing the breadth first search algorithm is 75.725%.

show abstract

“…Then, the kNN classifier is adopted for recognition and achieves an accuracy of 95.48%. Many other kNN-based OCR methods have also been proposed [23], [24].…”

Section: A Alphabetic Character Recognitionmentioning

confidence: 99%

CArDIS: A Swedish Historical Handwritten Character and Word Dataset

et al. 2022

View full text Add to dashboard Cite

This paper introduces a new publicly available image-based Swedish historical handwritten character and word dataset named Character Arkiv Digital Sweden (CArDIS) (https://cardisdataset.github. io/CARDIS/). The samples in CArDIS are collected from 64, 084 Swedish historical documents written by several anonymous priests between 1800 and 1900. The dataset contains 116, 000 Swedish alphabet images in RGB color space with 29 classes, whereas the word dataset contains 30, 000 image samples of ten popular Swedish names as well as 1, 000 region names in Sweden. To examine the performance of different machine learning classifiers on CArDIS dataset, three different experiments are conducted. In the first experiment, classifiers such as Support Vector Machine (SVM), Artificial Neural Networks (ANN), k-Nearest Neighbor (k-NN), Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and Random Forest (RF) are trained on existing character datasets which are Extended Modified National Institute of Standards and Technology (EMNIST), IAM and CVL and tested on CArDIS dataset. In the second and third experiments, the same classifiers as well as two pre-trained VGG-16 and VGG-19 classifiers are trained and tested on CArDIS character and word datasets. The experiments show that the machine learning methods trained on existing handwritten character datasets struggle to recognize characters efficiently on the CArDIS dataset, proving that characters in the CArDIS contain unique features and characteristics. Moreover, in the last two experiments, the deep learning-based classifiers provide the best recognition rates. INDEX TERMS Character and word recognition, Machine learning methods, Optical Character Recognition (OCR), Old handwritten style, Swedish handwritten character dataset, Swedish handwritten word dataset

show abstract

An analysis of optical character recognition implementation for ancient Batak characters using K-nearest neighbors principle

Cited by 8 publications

References 9 publications

A Prototypical Network-Based Approach for Low-Resource Font Typeface Feature Extraction and Utilization

A Prototypical Network-Based Approach for Low-Resource Font Typeface Feature Extraction and Utilization

One-shot learning Batak Toba character recognition using siamese neural network

CArDIS: A Swedish Historical Handwritten Character and Word Dataset

Contact Info

Product

Resources

About