A combination of features for symbol-independent writer identification in old music scores

The aim of writer identification is determining the writer of a piece of handwriting from a set of writers. In this paper, we present an architecture for writer identification in old handwritten music scores. Even though an important amount of music compositions contain handwritten text, the aim of our work is to use only music notation to determine the author. The main contribution is therefore the use of features extracted from graphical alphabets. Our proposal consists in combining the identification results of two different approaches, based on line and textural features. The steps of the ensemble architecture are the following. First of all, the music sheet is preprocessed for removing the staff lines. Then, music lines and texture images are generated for computing line features and textural features. Finally, the classification results are combined for identifying the writer. The proposed method has been tested on a database of old music scores from the seventeenth to nineteenth centuries, achieving a recognition rate of about 92% with 20 writers.     

A text-independent Persian writer identification based on feature relation graph (FRG)

The style of people's handwriting is a biometric feature that is used in person authentication. In this paper, we have proposed a text independent method for Persian writer identification. In the proposed method, pattern based features are extracted from data using Gabor and XGabor filter. The extracted features are represented for each person by using a graph that is called FRG (feature relation graph). This graph is constructed using relations between extracted features by employing a fuzzy method. The fuzzy method determines the similarity between features extracted from different handwritten instances of each person. In the identification phase, a graph similarity approach is employed to determine the similarity of the FRG generated from the test data and the FRGs generated by training data. The experimental results were satisfactory and the proposed method got about 100% accuracy on a dataset with 100 writers when enough training data was used. However, this method has been applied on Persian handwritings but we believe it can be extended on other languages especially in data representation and classification parts.     

WRITER IDENTIFICATION OF ARABIC TEXT USING STATISTICAL AND STRUCTURAL FEATURES

This article addresses writer identification of handwritten Arabic text. Several types of structural and statistical features were extracted from Arabic handwriting text. A novel approach was used to extract structural features that build on some of the main characteristics of the Arabic language. Connected component features for Arabic handwritten text as well as gradient distribution features, windowed gradient distribution features, contour chain code distribution features, and windowed contour chain code distribution features were extracted. A nearest neighbor (NN) classifier was used with the Euclidean distance measure. Data reduction algorithms (viz. principal component analysis [PCA], linear discriminant analysis [LDA], multiple discriminant analysis [MDA], multidimensional scaling [MDS], and forward/backward feature selection algorithm) were used. A database of 500 paragraphs handwritten in Arabic by 250 writers was used. The paragraphs used were randomly generated from a large corpus. NN provided the best accuracy in text-independent writer identification with top-1 result of 88.0%, top-5 result of 96.0%, and top-10 result of 98.5% for the first 100 writers. Extending the work to include all 250 writers and with the backward feature selection algorithm (using 54 out of 83 features), the system attained a top-1 result of 75.0%, top-5 result of 91.8%, and top-10 result of 95.4%.     

A writer identification system for on-line whiteboard data

In this paper we address the task of writer identification of on-line handwriting captured from a whiteboard. Different sets of features are extracted from the recorded data and used to train a text and language independent on-line writer identification system. The system is based on Gaussian mixture models (GMMs) which provide a powerful yet simple means of representing the distribution of the features extracted from the handwritten text. The training data of all writers are used to train a universal background model (UBM) from which a client specific model is obtained by adaptation. Different sets of features are described and evaluated in this work. The system is tested using text from 200 different writers. A writer identification rate of 98.56% on the paragraph and of 88.96% on the text line level is achieved.     

Text-independent writer identification and verification using textural and allographic features

The identification of a person on the basis of scanned images of handwriting is a useful biometric modality with application in forensic and historic document analysis and constitutes an exemplary study area within the research field of behavioral biometrics. We developed new and very effective techniques for automatic writer identification and verification that use probability distribution functions (PDFs) extracted from the handwriting images to characterize writer individuality. A defining property of our methods is that they are designed to be independent of the textual content of the handwritten samples. Our methods operate at two levels of analysis: the texture level and the character-shape (allograph) level. At the texture level, we use contour-based joint directional PDFs that encode orientation and curvature information to give an intimate characterization of individual handwriting style. In our analysis at the allograph level, the writer is considered to be characterized by a stochastic pattern generator of ink-trace fragments, or graphemes. The PDF of these simple shapes in a given handwriting sample is characteristic for the writer and is computed using a common shape codebook obtained by grapheme clustering. Combining multiple features (directional, grapheme, and run-length PDFs) yields increased writer identification and verification performance. The proposed methods are applicable to free-style handwriting (both cursive and isolated) and have practical feasibility, under the assumption that a few text lines of handwritten material are available in order to obtain reliable probability estimates     

Handwritten Chinese text line segmentation by clustering with distance metric learning

Separating text lines in unconstrained handwritten documents remains a challenge because the handwritten text lines are often un-uniformly skewed and curved, and the space between lines is not obvious. In this paper, we propose a novel text line segmentation algorithm based on minimal spanning tree (MST) clustering with distance metric learning. Given a distance metric, the connected components (CCs) of document image are grouped into a tree structure, from which text lines are extracted by dynamically cutting the edges using a new hypervolume reduction criterion and a straightness measure. By learning the distance metric in supervised learning on a dataset of pairs of CCs, the proposed algorithm is made robust to handle various documents with multi-skewed and curved text lines. In experiments on a database with 803 unconstrained handwritten Chinese document images containing a total of 8,169 lines, the proposed algorithm achieved a correct rate 98.02% of line detection, and compared favorably to other competitive algorithms.     

Partition selection with sparse autoencoders for content based image classification

Managing colossal image datasets with large dimensional hand-crafted features is no more feasible in most of the cases. Content based image classification (CBIC) of these large image datasets calls for the need of dimensionality reduction of features extracted for the purpose. This paper identifies the escalating challenges in the discussed domain and introduces a technique of feature dimension reduction by means of identifying region of interest in a given image with the use of reconstruction errors computed by sparse autoencoders. The automated process identifies the significant regions in an image for feature extraction. It not only improves the dimension of useful features but also contributes to increased classification results compared to earlier approaches. The reduction in number of one kind of features easily makes space for the inclusion of other features whose fusion facilitates improved classification performance compared to individual feature extraction techniques. Two different datasets, i.e. Wang dataset and Corel 5K dataset have been used for the experiments. State-of-the-art classifiers, i.e. Support Vector Machine and Extreme Learning Machine are used for CBIC. The proposed techniques are evaluated and compared in the context of both the classifiers and analysis of results suggests the appropriateness of the proposed methods for real time applications.

     

Open writer identification from offline handwritten signatures by jointing the one-class symbolic data analysis classifier and feature-dissimilarities

Usually, a large number of reference signatures are required for building the writing style model from offline handwritten signatures (OHSs). Moreover, the existing writer identification systems from OHSs are generally closed systems that require a retraining process when a new writer is added. This paper proposes an open writer identification system from OHSs, based on a new scheme of the one-class symbolic data analysis (OC-SDA) classifier, using few reference signatures. For generating more data, intra-class feature-dissimilarities, generated from curvelet transform, are introduced for building the symbolic representation model (SRM) associated with each writer. Feature-dissimilarities allow capturing more efficiently the intra-personnel variability produced naturally by a writer and, thus, increase the inter-personnel variability. Instead of using the mean and the standard deviation for building the OC-SDA model, intra-class feature-dissimilarities generated for each writer are modeled through a new weighted membership function, inspired from the real probability distribution of training intra-class feature-dissimilarities. The comparative analysis against the state-of-the-art works shows that the proposed OC-SDA classifier outperforms the existing classifiers on three public signature datasets GPDS-300, CEDAR-55 and MCYT-75, using only five reference signatures, achieving 98.31%, 98.06% and 99.89%, respectively, even when a combination of multiple classifiers is performed or even using learned features. Moreover, the evaluation of the proposed writer identification system in front of skilled forgeries shows its ability to detect also possible forged signatures in addition to the genuine ones.

     

An Unsupervised Writer Identification Based on Generating ClusterableEmbeddings

The writer identification system identifies individuals based on their handwriting is a frequent topic in biometric authentication and verification systems. Due to its importance, numerous studies have been conducted in various languages. Researchers have established several learning methods for writer identification including supervised and unsupervised learning. However, supervised methods require a large amount of annotation data, which is impossible in most scenarios. On the other hand, unsupervised writer identification methods may be limited and dependent on feature extraction that cannot provide the proper objectives to the architecture and be misinterpreted. This paper introduces an unsupervised writer identification system that analyzes the data and recognizes the writer based on the inter-feature relations of the data to resolve the uncertainty of the features. A pairwise architecture-based Autoembedder was applied to generate clusterable embeddings for handwritten text images. Furthermore, the trained baseline architecture generates the embedding of the data image, and the K-means algorithm is used to distinguish the embedding of individual writers. The proposed model utilized the IAM dataset for the experiment as it is inconsistent with contributions from the authors but is easily accessible for writer identification tasks. In addition, traditional evaluation metrics are used in the proposed model. Finally, the proposed model is compared with a few unsupervised models, and it outperformed the state-of-the-art deep convolutional architectures in recognizing writers based on unlabeled data.     

基于笔画曲率特征的笔迹鉴别方法

现有的手写汉字脱机笔迹鉴别方法存在只能针对特定字符或需要大量样本字符等问题,为此提出一种基于笔画曲率特征的笔迹鉴别方法。首先运用数学形态学对采集的笔迹图像进行预处理,在横、竖、撇、捺四个方向提取具有代表性的笔画骨架,然后对笔画骨架进行圆的重构,提取四个方向笔画圆的曲率作为特征值组成笔迹特征矩,根据待鉴别的笔迹特征矩与数据库中笔迹特征矩向量夹角相似性度量结果对样本做出判断。实验结果表明该文方法对于待鉴别样本字符的内容没有要求,样本字符数量要求低、应用范围广、鲁棒性强。     

Segmentation of text lines using multi-scale CNN from warped printed and handwritten document images

Paper documents are ideal sources of useful information and have a profound impact on every aspect of human lives. These documents may be printed or handwritten and contain information as combinations of texts, figures, tables, charts, etc. This paper proposes a method to segment text lines from both flatbed scanned/camera-captured heavily warped printed and handwritten documents. This work uses the concept of semantic segmentation with the help of a multi-scale convolutional neural network. The results of line segmentation using the proposed method outperform a number of similar proposals already reported in the literature. The performance and efficacy of the proposed method have been corroborated by the test result on a variety of publicly available datasets, including ICDAR, Alireza, IUPR, cBAD, Tobacco-800, IAM, and our dataset.

     

Wavelet-based gender detection on off-line handwritten documents using probabilistic finite state automata

Detection of gender from handwriting of an individual presents an interesting research problem with applications in forensic document examination, writer identification and psychological studies. This paper presents an effective technique to predict the gender of an individual from off-line images of handwriting. The proposed technique relies on a global approach that considers writing images as textures. Each handwritten image is converted into a textur\ed image which is decomposed into a series of wavelet sub-bands at a number of levels. The wavelet sub-bands are then extended into data sequences. Each data sequence is quantized to produce a probabilistic finite state automata (PFSA) that generates feature vectors. These features are used to train two classifiers, artificial neural network and support vector machine to discriminate between male and female writings. The performance of the proposed system was evaluated on two databases, QUWI and MSHD, within a number of challenging experimental scenarios and realized classification rates of up to 80%. The experimental results show the superiority of the proposed technique over existing techniques in terms of classification rates.     

Choice of a classifier,based on properties of a dataset: case study-speech emotion recognition

In this paper, the process of selecting a classifier based on the properties of dataset is designed since it is very difficult to experiment the data on n—number of classifiers. As a case study speech emotion recognition is considered. Different combinations of spectral and prosodic features relevant to emotions are explored. The best subset of the chosen set of features is recommended for each of the classifiers based on the properties of chosen dataset. Various statistical tests have been used to estimate the properties of dataset. The nature of dataset gives an idea to select the relevant classifier. To make it more precise, three other clustering and classification techniques such as K-means clustering, vector quantization and artificial neural networks are used for experimentation and results are compared with the selected classifier. Prosodic features like pitch, intensity, jitter, shimmer, spectral features such as mel frequency cepstral coefficients (MFCCs) and formants are considered in this work. Statistical parameters of prosody such as minimum, maximum, mean (\(\mu\)) and standard deviation (\(\sigma\)) are extracted from speech and combined with basic spectral (MFCCs) features to get better performance. Five basic emotions namely anger, fear, happiness, neutral and sadness are considered. For analysing the performance of different datasets on different classifiers, content and speaker independent emotional data is used, collected from Telugu movies. Mean opinion score of fifty users is collected to label the emotional data. To make it more accurate, one of the benchmark IIT-Kharagpur emotional database is used to generalize the conclusions.     

KHATT: An open Arabic offline handwritten text database

A comprehensive Arabic handwritten text database is an essential resource for Arabic handwritten text recognition research. This is especially true due to the lack of such database for Arabic handwritten text. In this paper, we report our comprehensive Arabic offline Handwritten Text database (KHATT) consisting of 1000 handwritten forms written by 1000 distinct writers from different countries. The forms were scanned at 200, 300, and 600 dpi resolutions. The database contains 2000 randomly selected paragraphs from 46 sources, 2000 minimal text paragraph covering all the shapes of Arabic characters, and optionally written paragraphs on open subjects. The 2000 random text paragraphs consist of 9327 lines. The database forms were randomly divided into 70%, 15%, and 15% sets for training, testing, and verification, respectively. This enables researchers to use the database and compare their results. A formal verification procedure is implemented to align the handwritten text with its ground truth at the form, paragraph and line levels. The verified ground truth database contains meta-data describing the written text at the page, paragraph, and line levels in text and XML formats. Tools to extract paragraphs from pages and segment paragraphs into lines are developed. In addition we are presenting our experimental results on the database using two classifiers, viz. Hidden Markov Models (HMM) and our novel syntactic classifier.     

Automatic writer identification framework for online handwritten documents using character prototypes

This paper proposes an automatic text-independent writer identification framework that integrates an industrial handwriting recognition system, which is used to perform an automatic segmentation of an online handwritten document at the character level. Subsequently, a fuzzy c-means approach is adopted to estimate statistical distributions of character prototypes on an alphabet basis. These distributions model the unique handwriting styles of the writers. The proposed system attained an accuracy of 99.2% when retrieved from a database of 120 writers. The only limitation is that a minimum length of text needs to be present in the document in order for sufficient accuracy to be achieved. We have found that this minimum length of text is about 160 characters or approximately equivalent to 3 lines of text. In addition, the discriminative power of different alphabets on the accuracy is also reported.     

回归—聚类联合框架下的手写文本行提取

目的 手写文本行提取是文档图像处理中的重要基础步骤,对于无约束手写文本图像,文本行都会有不同程度的倾斜、弯曲、交叉、粘连等问题。利用传统的几何分割或聚类的方法往往无法保证文本行边缘的精确分割。针对这些问题提出一种基于文本行回归-聚类联合框架的手写文本行提取方法。方法 首先,采用各向异性高斯滤波器组对图像进行多尺度、多方向分析,利用拖尾效应检测脊形结构提取文本行主体区域,并对其骨架化得到文本行回归模型。然后,以连通域为基本图像单元建立超像素表示,为实现超像素的聚类,建立了像素-超像素-文本行关联层级随机场模型,利用能量函数优化的方法实现超像素的聚类与所属文本行标注。在此基础上,检测出所有的行间粘连字符块,采用基于回归线的k-means聚类算法由回归模型引导粘连字符像素聚类,实现粘连字符分割与所属文本行标注。最后,利用文本行标签开关实现了文本行像素的操控显示与定向提取,而不再需要几何分割。结果 在HIT-MW脱机手写中文文档数据集上进行文本行提取测试,检测率DR为99.83%,识别准确率RA为99.92%。结论 实验表明,提出的文本行回归-聚类联合分析框架相比于传统的分段投影分析、最小生成树聚类、Seam Carving等方法提高了文本行边缘的可控性与分割精度。在高效手写文本行提取的同时,最大程度地避免了相邻文本行的干扰,具有较高的准确率和鲁棒性。     

Novel mislabeled training data detection algorithm

As a kind of noise, mislabeled training data exist in many applications. Because of their negative effects on learning, many filter techniques have been proposed to identify and eliminate them. Ensemble learning-based filter (EnFilter) is the most widely used filter which employs ensemble classifiers. In EnFilter, first the noisy training dataset is divided into several subsets. Each noisy subset is then checked by the multiple classifiers which are trained based on other noisy subsets. It is noted that since the training data used to train multiple classifiers are noisy, the quality of these classifiers cannot be guaranteed, which might generate poor noise identification result. This problem is more serious when the noise ratio in the training dataset is high. To solve this problem, a straightforward but effective approach is proposed in this work. Instead of using noisy data to train the classifiers, nearly noise-free (NNF) data are used since they are supposed to train more reliable classifiers. To this end, a novel NNF data extraction approach is also proposed. Experimental results on a set of benchmark datasets illustrate the utility of our proposed approach.