FragNet: Writer Identification Using Deep Fragment Networks

He, Sheng; Schomaker, Lambert

doi:10.1109/tifs.2020.2981236

Cited by 64 publications

(32 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For each backbone (ResNet+TGCNN, MobileNet+TGCNN, Xcep-tion+TGCNN, and DenseNet+TGCNN), the architectures directly receive the inputs from the thresholded Gabor filter. Architectures specially designed for writer recognition (DeepWriter [65] and FragNet [5]) are trained using wordlevel images. Every architecture is pre-trained using a small subset of ImageNet [66] dataset.…”

Section: B Competitive Architecturesmentioning

confidence: 99%

A Thresholded Gabor-CNN Based Writer Identification System for Indic Scripts

Mridha¹,

Ohi

Shin

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Writer identification is the procedure of identifying individuals from handwriting. Writer identification is a common interest in biometric authentication and verification systems, and numerous studies are available for English, Chinese, Arabic, and Persian specific handwriting. This paper introduces a supervised offline Indic script writer identification system that can identify individuals using less than a single page of handwriting. A lightweight Convolutional Neural Network (CNN) architecture fused with non-trainable Gabor filters is used as an identification model that can recognize writers from scarce data. For the experiment, we used BanglaWriting dataset, which is openly available for Bengali writing and writer recognition. Further, we added Devanagari and Telugu datasets for evaluation. The overall evaluation shows that the proposed thresholded Gabor-based CNN architecture performs superior to numerous deep CNN architectures for Indic writer recognition.

show abstract

Section: B Competitive Architecturesmentioning

confidence: 99%

A Thresholded Gabor-CNN Based Writer Identification System for Indic Scripts

Mridha¹,

Ohi

Shin

et al. 2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…He and Schomaker have recently proposed two deep architectures [32,33]. The former one represents the handwriting which contains two implicit and explicit information where the explicit information refers to the length, number of characters, and lexical contents of single words.…”

Section: Literature Reviewmentioning

confidence: 99%

User Authentication Based on Handwriting Analysis of Pen-Tablet Sensor Data Using Optimal Feature Selection Model

et al. 2021

View full text Add to dashboard Cite

Handwriting analysis is playing an important role in user authentication or online writer identification for more than a decade. It has a significant role in different applications such as e-security, signature biometrics, e-health, gesture analysis, diagnosis system of Parkinson’s disease, Attention-deficit/hyperactivity disorders, analysis of vulnerable people (stressed, elderly, or drugged), prediction of gender, handedness and so on. Classical authentication systems are image-based, text-dependent, and password or fingerprint-based where the former one has the risk of information leakage. Alternatively, image processing and pattern-analysis-based systems are vulnerable to camera attributes, camera frames, light effect, and the quality of the image or pattern. Thus, in this paper, we concentrate on real-time and context-free handwriting data analysis for robust user authentication systems using digital pen-tablet sensor data. Most of the state-of-the-art authentication models show suboptimal performance for improper features. This research proposed a robust and efficient user identification system using an optimal feature selection technique based on the features from the sensor’s signal of pen and tablet devices. The proposed system includes more genuine and accurate numerical data which are used for features extraction model based on both the kinematic and statistical features of individual handwritings. Sensor data of digital pen-tablet devices generate high dimensional feature vectors for user identification. However, all the features do not play equal contribution to identify a user. Hence, to find out the optimal features, we utilized a hybrid feature selection model. Extracted features are then fed to the popular machine learning (ML) algorithms to generate a nonlinear classifier through training and testing phases. The experimental result analysis shows that the proposed model achieves more accurate and satisfactory results which ensure the practicality of our system for user identification with low computational cost.

show abstract

“…He and Schomaker [39], in a later work, proposed Frag-Net. This deep neural network entails two pathways: a feature pyramid first used to extract feature maps and a fragment pathway based on fragments extracted from the input image and the feature maps from the feature pyramid.…”

Section: Deep Learning Approachesmentioning

confidence: 99%

“…Although relatively recent works on the writer identification field utilize Artificial Intelligence techniques [7,34,36,[38][39][40]42], a choice was made to use traditional methods only for several reasons.…”

Section: Motivation Objectives and Assumptionsmentioning

confidence: 99%

“…Some researchers [36] try to overcome those limitations using another Artificial Intelligence technique to generate thousands of samples per writer. Other researchers [39,40] utilize the annotations available on Data sets to split entire pages into words or proceed with word segmentation techniques and then try to achieve writer identification on the word level.…”

Section: Motivation Objectives and Assumptionsmentioning

confidence: 99%

See 1 more Smart Citation