Handwritten word recognition using lottery ticket hypothesis based pruned CNN model: a new benchmark on CMATERdb2.1.2

Malakar, Samir; Paul, Sayantan; Kundu, Soumyadeep; Bhowmik, Showmik; Sarkar, Ram; Nasipuri, Mita

doi:10.1007/s00521-020-04872-0

Cited by 13 publications

(2 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this context, it is worth mentioning that the existing OCR systems do not perform satisfactorily for handwritten texts due to the extreme writing style variation of the individuals [3,4]. Besides, segmentation ambiguity becomes a prevalent issue when handwriting is cursive [5,6]. Another notable problem with OCR-based solutions is that it tries to convert each word present in the document which is often unnecessary and makes the problem time-consuming and errorprone [7,8].…”

Section: Introductionmentioning

confidence: 99%

Z-Transform-Based Profile Matching to Develop a Learning-Free Keyword Spotting Method for Handwritten Document Images

Banerjee

Bhowal²,

Malakar³

et al. 2022

Int J Comput Intell Syst

Self Cite

View full text Add to dashboard Cite

For easy accessibility of the information from the digitized document images, optical character recognition (OCR)-based software can be used. But in the case of handwritten documents, the performance of the state-of-the-art OCR systems is not satisfactory owing to the complexity of the unconstrained handwriting. Hence, research affinity comes up with an alternative solution for this problem called keyword spotting (KWS) which is much more practical than an OCR-based solution. This work proposes a novel learning-free KWS method that can be applied to a heterogeneous collection of handwritten documents. In this work, we introduce a new way of profile matching to compare the query word profiles (i.e., both upper and lower) with the target words’ profiles. At first, both query and target words are binarized, and then two profiles from each such word are generated. Next, we formulate rules to filter out the irrelevant words concerning the query word and obtain the probable candidate query (i.e., target) words. Then we compare the profiles of the query and candidate query words in the Z-transform domain using the condition of resonance for the damped oscillator. However, before the match, we perform an affine transformation on the Bezier curve representation of the profiles of the candidate query words to reduce the effects like scaling, rotation, and shearing which might occur due to the variant writing styles of individuals. The proposed method achieves satisfactory performance compared to state-of-the-art learning-free methods when applied to four publicly available standard datasets namely ICFHR 2014 H-KWS competition Modern, IAM, ICFHR 2016 H-KWS competition Botany and ICFHR 2016 H-KWS competition Konzilsprotokolle datasets.

show abstract

Section: Introductionmentioning

confidence: 99%

Z-Transform-Based Profile Matching to Develop a Learning-Free Keyword Spotting Method for Handwritten Document Images

Banerjee

Bhowal²,

Malakar³

et al. 2022

Int J Comput Intell Syst

Self Cite

View full text Add to dashboard Cite

show abstract

“…The accuracy rates for the training set and test set were 99.98% and 99.39% respectively. In addition to the aforementioned approaches, several recent studies have explored deep learning-based models for writer recognition in the Bangla script [35][36][37]. In [38] Adak et al worked on page-level Bangla script with auto crafted features.…”

Section: Related Workmentioning

confidence: 99%

An integrated approach towards improvement of writer verification performance using page level, line level and word level data on Bangla handwritten document

Paul

Dutta

Sarkar

et al. 2023

Preprint

View full text Add to dashboard Cite

Handwriting analysis presents a formidable challenge, especially when dealing with languages like Bangla that lack extensive resources. Previous research on Bangla handwritten document recognition has focused on various levels, including page, block, line, and word analysis. However, the improvement of page-level recognition in this context remains an unexplored area. This paper aims to fill this gap by proposing an integrated approach that combines techniques at the page level, line level, and word level for writer verification. We have developed a novel method that significantly improves the performance of page-level writer verification, utilizing our newly created dataset, JUDVLP-BLWVdb. By utilizing the widely recognized ensemble technique of majority voting, we combine three classifiers (Support Vector Machine, Multilayer Perceptron, and Simple Logistic). Our proposed method achieves a remarkable increase in page level writer verification accuracy, improving by it nearly 12% to 97.62% across 101 different writers. Furthermore, we compare the results of our dataset with four state-of-the-art writer verification approaches. Additionally, we explore the application of deep learning-based approaches, including VGG16, ResNet34, and AlexNet, using our dataset.

show abstract

Comparative Analysis of Performances of Convolutional Neural Networks for Image Classification Tasks

Alfa¹,

Misra

Yusuf

et al. 2023

Lecture Notes in Electrical Engineering

View full text Add to dashboard Cite

Handwritten word recognition using lottery ticket hypothesis based pruned CNN model: a new benchmark on CMATERdb2.1.2

Cited by 13 publications

References 23 publications

Z-Transform-Based Profile Matching to Develop a Learning-Free Keyword Spotting Method for Handwritten Document Images

Z-Transform-Based Profile Matching to Develop a Learning-Free Keyword Spotting Method for Handwritten Document Images

An integrated approach towards improvement of writer verification performance using page level, line level and word level data on Bangla handwritten document

Comparative Analysis of Performances of Convolutional Neural Networks for Image Classification Tasks

Contact Info

Product

Resources

About