Automatic Gender Classification from Handwritten Images: A Case Study

Rabaev, Irina; Litvak, Marina; Asulin, Sean; Tabibi, Or Haim

doi:10.1007/978-3-030-89131-2_30

Cited by 6 publications

(12 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When applied to English and Arabic subsets, the ResNet network had a better accuracy rate of 66.06% and 72%, respectively. The results of ResNet in terms of accuracy outperform the results obtained by Xue et al[13] and Rabaev et al[27], both of which used deep learning in their methodologies. The two CNN architectures were also evaluated using ROC and AUC on the ICDAR2013 dataset.…”

mentioning

confidence: 54%

Transfer Learning to Detect Age From Handwriting

AL-Qawasmeh¹,

Suen²

2022

AAIML

View full text Add to dashboard Cite

Handwriting analysis is the science of determining an individual’s personality from his or her handwriting by assessing features such as slant, pen pressure, word spacing, and other factors. Handwriting analysis has a wide range of uses and applications, including dating and socialising, roommates and landlords, business and professional, employee hiring, and human resources. This study used the ResNet and GoogleNet CNN architectures as fixed feature extractors from handwriting samples. SVM was used to classify the writer’s gender and age based on the extracted features. We built an Arabic dataset named FSHS to analyse and test the proposed system. In the gender detection system, applying the automatic feature extraction method to the FSHS dataset produced accuracy rates of 84.9% and 82.2% using ResNet and GoogleNet, respectively. While the age detection system using the automatic feature extraction method achieved accuracy rates of 69.7% and 61.1% using ResNet and GoogleNet, respectively

show abstract

mentioning

confidence: 54%

Transfer Learning to Detect Age From Handwriting

AL-Qawasmeh¹,

Suen²

2022

AAIML

View full text Add to dashboard Cite

show abstract

“…Deep models based on Convolutional Neural Networks started to appear in gender classification works around 2018. Deep neural networks were applied as feature extractors [21], and also end-to-end pipelines, including both feature selection and classification layers [8,32,33]. The main advantage of deep networks is their ability to learn features automatically without manual engineering.…”

Section: Gender Classificationmentioning

confidence: 99%

“…The main advantage of deep networks is their ability to learn features automatically without manual engineering. In addition, CNNs have been shown to be on par or even outperforming other classifiers on gender classification task [8,33,34]. Due to their benefits in terms of performance and usability, deep networks have recently emerged as a leader in various computer vision applications, including handwriting analysis.…”

Section: Gender Classificationmentioning

confidence: 99%

“…Based on the success of the bilinear convolutional neural network (B-CNN) for finegrained classification, we offer a novel implementation of a B-CNN [7] using ResNet instead of VGG blocks. In our previous study [8], we compared multiple deep CNNs on the gender classification task, where VGG demonstrated poor performance. Therefore, in this work, we decided to replace VGG blocks with ResNet, which showed better performance and was relatively easy to integrate within B-CNN.…”

Section: Introductionmentioning

confidence: 99%

“…The main contributions of our work are summarized as follows: (1) We modify and apply B-CNN for document image classification; (2) This is the first attempt to apply deep learning models for age classification from handwritten document images; (3) We substitute VGG blocks in the B-CNN architecture by ResNet due to the VGG poor performance in our previous study on gender prediction from handwriting [8]; (4) We perform thorough experiments on three benchmark datasets consisting of handwritten documents in three different languages: Arabic, English, and Hebrew. We show that when enough training data are available, B-ResNet outperforms other methods.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automatic Gender and Age Classification from Offline Handwriting with Bilinear ResNet

Rabaev

Alkoran

Wattad

et al. 2022

Sensors

Self Cite

View full text Add to dashboard Cite

This work focuses on automatic gender and age prediction tasks from handwritten documents. This problem is of interest in a variety of fields, such as historical document analysis and forensic investigations. The challenge for automatic gender and age classification can be demonstrated by the relatively low performances of the existing methods. In addition, despite the success of CNN for gender classification, deep neural networks were never applied for age classification. The published works in this area mostly concentrate on English and Arabic languages. In addition to Arabic and English, this work also considers Hebrew, which was much less studied. Following the success of bilinear Convolutional Neural Network (B-CNN) for fine-grained classification, we propose a novel implementation of a B-CNN with ResNet blocks. To our knowledge, this is the first time the bilinear CNN is applied for writer demographics classification. In particular, this is the first attempt to apply a deep neural network for the age classification. We perform experiments on documents from three benchmark datasets written in three different languages and provide a thorough comparison with the results reported in the literature. B-ResNet was top-ranked in all tasks. In particular, B-ResNet outperformed other models on KHATT and QUWI datasets on gender classification.

show abstract