Local gradient full-scale transform patterns based off-line text-independent writer identification

Chahi, Abderrazak; Merabet, Youssef El; Ruichek, Yassine; Touahni, Raja

doi:10.1016/j.asoc.2020.106277

Cited by 19 publications

(7 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For this reason, we exclude one of the two fixed text paragraphs as mentioned in section 4.1. All methods in [ 14 , 58 , 59 ] which outperformed our proposed system on KHATT dataset do not refer to which group (text-dependent or text-independent) their systems belong, however, their dataset setup does not exclude the repeated text (fixed text paragraphs) which in turn leads to higher identification rate as reported by [ 30 ]. Furthermore, this approach-wise difference leads to the usage of different handwritten samples per writer which does not guarantee fair performance comparison [ 61 ].…”

Section: Resultsmentioning

confidence: 99%

“…The authors reported that the proposed system achieves 94.89% on IFN/ENIT and 89.54% on IAM using the complete set of writers from the two datasets. Chahi et al [ 30 ] proposed an offline handwriting writer identification system in which the learning method exploits small local regions of the handwritten document. The document is segmented into connected components, which in turn are fed into a cross multiscale locally encoded gradient patterns (CLGP) operator to compute a feature vector.…”

Section: Related Workmentioning

confidence: 99%

“…Auto learned methods, sometimes called model-based methods rely on the usage of deep learning techniques to extract automatic features learned by deep models. The main drawback of deep models is the need for enormous and massive labeled data to learn, which is not the case in all benchmarks that are usually used in the writer identification task [ 30 ]. According to [ 34 ], another important drawback of deep techniques is the difficulty associated with selecting the best values for a large number of parameters.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Offline text-independent writer identification using a codebook with structural features

2023

View full text Add to dashboard Cite

Understanding handwritten documents is a vital and challenging problem that attracts many researchers in the fields of forensic and authentication science. This paper presents an offline system for text-independent writer identification of handwritten documents. The system extracts a handwritten connected component contour, which in turn is divided into segments of specific length. The system utilizes the concept of a bag of features in the writer recognition domain and considers handwritten contour segments to extract two conceptually simple and effective structural features. These features are the contour point curve angle and the CONtour point CONcavity/CONvexity. The system uses the proposed features to train a k-means clustering algorithm to construct a codebook of size K. The method then uses occurrence histograms of the extracted features in the codebook to create a final feature vector for each handwritten document. The effectiveness of the proposed features is evaluated in the writer identification domain using two widely used classification methods: the nearest neighbor and the support vector machine techniques. The proposed writer identification is evaluated on two large and public datasets from different language domains, the Arabic KHATT and English IAM datasets. The experimental results show that the proposed system outperforms state-of-the-art methods on the IAM dataset and provides competitive results on the KHATT dataset with respect to the identification rate.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Offline text-independent writer identification using a codebook with structural features

2023

View full text Add to dashboard Cite

show abstract

“…In addition, the authors [15] proposed a new method called (DLS_CNN) for writer recognition so, in this research used the combination between neural network (NN) with line segmentation. On the other hand in Chahi et al [16], the authors proposed a new algorithm called LSTP but at the classification step base on NN achieved Hamming distance.…”

Section: Literature Reviewmentioning

confidence: 99%

Offline Writer Recognition for Kurdish Handwritten Text Document Based on Proposed Codebook

Mohammed

Ahmed²

2021

UHD J SCI TECH

View full text Add to dashboard Cite

Handwritten text recognition has been an ongoing attractive task to research in the field of document analysis and recognition with applications in handwriting forensics, paleography, document examination, and handwriting recognition. In the present research, an automatic method of writer recognition is presented using digitized images of unconstrained texts. Despite the increasing efforts by prior literature on the different methods used for the same purpose, such methods performance, particularly their accuracy, has not been promising, leaving plenty of room for improvements. This method made use of codebook-based writer characterization, with each writing sample represented by a group of computed features from a primary and secondary codebook. The writings were then represented through the computation of the probability of codebook patterns occurrence, and the probability distribution was employed for each writer’s characterization. Writer identification process involved comparing two writings through the computation of the distances between their respective probability distribution. The study carried out experiments to determine the performance of the implemented method in light of rates of identification with the help of standard datasets, namely, KRDOH and IAM, the former being the most current and largest Kurdish handwritten datasets with 1076 writers, and the latter being a dataset containing 650 writers. The outcome of the experiments was promising with a rate of identification of 94.3%, with the proposed method outperforming the state-of-the-art methods by 2–3%.

show abstract

“…Forensic writer identification refers to the task of identifying a specific writer of a piece of handwriting, which has potential applications in forensic document examination [1] and historical manuscript analysis [2,3,4]. The classical methods [5,6,7,8] use shape or texture features of handwritten text to recognize the writer, which requires a large amount of image information per sample in order to obtain a statistically reliable feature vector [2,6]. Therefore, most studies focus on writer identification using page-level document images which contain several paragraphs or sentences.…”

Section: Introductionmentioning

confidence: 99%

GR-RNN: Global-Context Residual Recurrent Neural Networks for Writer Identification

Schomaker

2021

Preprint

View full text Add to dashboard Cite

This paper presents an end-to-end neural network system to identify writers through handwritten word images, which jointly integrates global-context information and a sequence of local fragment-based features. The global-context information is extracted from the tail of the neural network by a global average pooling step. The sequence of local and fragment-based features is extracted from a low-level deep feature map which contains subtle information about the handwriting style. The spatial relationship between the sequence of fragments is modeled by the recurrent neural network (RNN) to strengthen the discriminative ability of the local fragment features. We leverage the complementary information between the global-context and local fragments, resulting in the proposed global-context residual recurrent neural network (GR-RNN) method.The proposed method is evaluated on four public data sets and experimental results demonstrate that it can provide state-of-the-art performance. In addition, the neural networks trained on gray-scale images provide better results than neural networks trained on binarized and contour images, indicating that texture information plays an important role for writer identification.

show abstract

Local gradient full-scale transform patterns based off-line text-independent writer identification

Cited by 19 publications

References 45 publications

Offline text-independent writer identification using a codebook with structural features

Offline text-independent writer identification using a codebook with structural features

Offline Writer Recognition for Kurdish Handwritten Text Document Based on Proposed Codebook

GR-RNN: Global-Context Residual Recurrent Neural Networks for Writer Identification

Contact Info

Product

Resources

About