Co-occurrence Features for Writer Identification

He, Sheng; Schomaker, Lambert

doi:10.1109/icfhr.2016.0027

Cited by 10 publications

(10 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Requires careful design or selection of the parameter values of filters. [1,2,27,28] Extracts the writing style information based on ink trace by edge or contour angles. The feature vector is the joint distribution of angles on each position of ink trace.…”

Section: References Featuresmentioning

confidence: 99%

“…Filter-based features, such as Gabor [24], XGabor [25] and oriented Basic Image Feature Columns (oBIF Columns) [26], have also been studied. Some features can be extracted from the contours of the ink trace, such as Hinge-based features [1,2,27,28], which extract the slant property of characters alongside other information, such as stroke width [2] and curvature information [3]. Other features, such as symbolic representation [35] and kadjacent segments (kAS) [36,37] are also used for writer identification.…”

Section: References Featuresmentioning

confidence: 99%

See 1 more Smart Citation

Deep adaptive learning for writer identification based on single handwritten word images

Schomaker

2019

Pattern Recognition

Self Cite

View full text Add to dashboard Cite

There are two types of information in each handwritten word image: explicit information which can be easily read or derived directly, such as lexical content or word length, and implicit attributes such as the author's identity. Whether features learned by a neural network for one task can be used for another task remains an open question. In this paper, we present a deep adaptive learning method for writer identification based on single-word images using multi-task learning. An auxiliary task is added to the training process to enforce the emergence of reusable features. Our proposed method transfers the benefits of the learned features of a convolutional neural network from an auxiliary task such as explicit content recognition to the main task of writer identification in a single procedure. Specifically, we propose a new adaptive convolutional layer to exploit the learned deep features. A multi-task neural network with one or several adaptive convolutional layers is trained end-to-end, to exploit robust generic features for a specific main task, i.e., writer identification. Three auxiliary tasks, corresponding to three explicit attributes of handwritten word images (lexical content, word length and character attributes), are evaluated. Experimental results on two benchmark datasets show that the proposed deep adaptive learning method can improve the performance of writer identification based on singleword images, compared to non-adaptive and simple linear-adaptive approaches.

show abstract

Section: References Featuresmentioning

confidence: 99%

Section: References Featuresmentioning

confidence: 99%

Deep adaptive learning for writer identification based on single handwritten word images

Schomaker

2019

Pattern Recognition

Self Cite

View full text Add to dashboard Cite

show abstract

“…The CoHinge feature [13] is the joint distribution of the Hinge kernel on two different pixels of writing contours based on spatial joint feature distribution described in [14]. First, we extract ink contours from the binarized image.…”

Section: Groningen: Alice University Of Groningen the Netherlandmentioning

confidence: 99%

ICDAR2017 Competition on Historical Document Writer Identification (Historical-WI)

Fiel

Kleber

Diem

et al. 2017

2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR)

View full text Add to dashboard Cite

The ICDAR 2017 Competition on Historical Document Writer Identification is dedicated to record the most recent advances made in the field of writer identification. The goal of the writer identification task is the retrieval of pages, which have been written by the same author. The test dataset used in this competition consists of 3600 handwritten pages originating from 13 th to 20 th century. It contains manuscripts from 720 different writers where each writer contributed five pages. This paper describes the dataset, as well as the details of the competition. Five different institutions submitted six methods which were ranked using identification and retrieval metrics. The paper describes the competition details including the dataset, the evaluation measures used as well as a short description of each submitted method.

show abstract

“…Bulacu et al [16] introduced the joint probability distribution of the angle combination of two edge fragments for feature extraction. In order to gain location‐specific information, features were computed on the top and the bottom halves of text lines separately, and then, the results were fused [17]. They also extracted direction features from contours of connected components more efficiently and more rapidly by the contour‐hinge method [10, 18].…”

Section: Related Workmentioning

confidence: 99%

Writer identification with n ‐tuple direction feature from contour

et al. 2020

View full text Add to dashboard Cite

This study introduces an effective solution for text‐independent writer identification by generalising contour‐hinge feature, which is called n‐tuple direction feature. For extracting n‐tuple direction feature, the authors first obtain all contours from connected components, then n + 1 points are considered on the contour with a certain distance apart, and next, the directions of the fragments connecting two successive points are computed. The n + 1 points move on the contour and the n‐dimensional histogram of directions is computed. The proposed method is evaluated on large Farsi and English databases. A correct writer identification rate of 92.2% for English handwritings from 900 persons and 97.7% for Farsi handwritings from 600 persons are achieved. Comparison between the proposed method and other studies shows the promising performance and superiority of the proposed method.

show abstract

Co-occurrence Features for Writer Identification

Cited by 10 publications

References 39 publications

Deep adaptive learning for writer identification based on single handwritten word images

Deep adaptive learning for writer identification based on single handwritten word images

ICDAR2017 Competition on Historical Document Writer Identification (Historical-WI)

Writer identification with n ‐tuple direction feature from contour

Contact Info

Product

Resources

About