Search

Handwriting is a component of the complex language that came about late in the history of mankind and which develops late in human beings. Numerous works have raised changes in both the graphic and kinematic characteristics of writing. Although, age does not modify the lexical and syntactic parameters of language, it can however modify its spatial structure, especially pressure and speed. Many neurodegenerative pathologies, especially Alzheimer's disease, are characterized by a progressive disorganization of writing. Depending on the cognitive stage of the dementia, the graphic gesture deteriorates as does the spatial construction. Objective: Our study aims at assessing the characteristics of Arabic writing in a healthy Moroccan population and to compare it to people with mild to moderate Alzheimer's disease. Our objective is to help health professionals detect early cognitive deterioration in neurodegenerative diseases by analyzing the graphic gesture. Handwriting is captured on a graphic tablet (WACOM) and is analyzed "online" as a sequence of acquired signals (position, pressure, speed and pen inclination) in Moroccan patients with mild to moderate Alzheimer's disease and these were compared to those of normal volunteers. We performed a first analysis of the results from 18 Alzheimer's patients compared to 18 control subjects. The results reveal differences between the control and Alzheimer's groups. AD subjects had lower speeds and accelerations compared to the control subjects. The time spent on paper and in the air was significantly greater in the AD subjects. This preliminary analysis of the results allowed us to identify distinguishing characteristics through the analysis of different handwriting parameters in order to identify the two groups studied.

Parkinson’s disease (PD) is a progressive movement disorder characterized by tremors at rest, bradykinesia, and stifness. The alteration of handwriting (HW) faculties is one of the earliest motor symptoms in PD patients. This characteristic can be exploited to develop an automatic aid system for early detection of this pathologie. This article aims to assess the importance of diacritics and punctuation marks (DPM) in the PD patients and healthy controls (HCs) discrimination problem, by comparing the classification results obtained from three components: text carrying DPM, text without DPM, as well as only DPM. This work includes the Arabic manuscripts of 31 PD patients and 31 HCs. Furthermore, kinematic, mechanic, and inclination features were calculated for each component. Then, Adaboost models have been constructed on different feature sets, as well as on reduced sets formed in incremental manner using mRMR ranked-feature selection method. From the obtained results, it was concluded that the separating power of HW features in the classification problem of PD patients and HCs is present in all components of the Arabic text, but in varying degrees of importance. Despite the simple graphical nature of DPM, they are carrying of relevant diagnostic information, and effectively contributing to the improvement of PD detection performance. The highest accuracy of 93.54% was achieved for this component.

Parkinson’s disease (PD) is a neurodegenerative disorder that affects a person’s movement. As the early diagnosis of the disease is crucial, the main aim of this work is to implement an online analysis system of patients’ handwriting, through computer vision and signal processing techniques, using the database collected in the neurology department of the University Hospital Center Hassan II in Fez. For this, we studied the handwriting tests on a WACOM graphic tablet to retrieve the spatiotemporal data (position, pressure and angles of inclination), for each point (P(n)) of the trajectory. The features vector was obtained basing on five types of features: (a) Kinematic features related to the dynamics of spiral design, (b) Mechanical based on the pressure exerted on the writing surface, (c) Inclination angles, (d) Spatial interrelation feature and (e) Pen-Up. The used classification and clustering algorithms are respectively the Hoeffding tree and the FarthestFirst clusters. We observed coherence between the classification results and the clustering ones, thus the results being encouraging and promising with a recognition rate of 98.36%

The similarity or the distance measure have been used widely to calculate the similarity or dissimilarity between vector sequences, where the document images similarity is known as the domain that dealing with image information and both similarity/distance has been an important role for matching and pattern recognition. There are several types of similarity measure, we cover in this paper the survey of various distance measures used in the images matching and we explain the limitations associated with the existing distances. Then, we introduce the concept of the floating distance which describes the variation of the threshold’s selection for each word in decision making process, based on a combination of Linear Regression and cosine distance. Experiments are carried out on a handwritten Arabic image documents of Gallica library. These experiments show that the proposed floating distance outperforms the traditional distance in word spotting system.

Parkinson’s disease (PD) is one of the most common neurodegenerative diseases affecting a large population worldwide. Parkinson’s disease is characterized by rigidity, slowness of movement and tremors at rest, these syndromes are frequently manifested in the deterioration of handwriting. The aim of this article is to perform online Arabic handwriting analysis for two types of tasks, TASK 1: copying arabic imposed text and TASK 2: writing arabic desired text. A novel method of handwriting selection features is proposed to obtain the relevant features to efficiently identify subjects with PD, based on Recursive Feature Elimination with Cross-Validation (RFECV), three different RFE estimators were compared: Support Vector Machine, Decision Trees and Random Forest, the selected features have been fed to the same classifiers above to determine the best classifier for predicting Parkinson’s disease. Result: An accuracy of 94.4% was obtained using SVM with Linear kernel, based on 55 features selected using RFE-SVM(Linear) for TASK 1, for TASK 2 an accuracy of 93.7% was obtained using SVM with RBF kernel, based only in 7 features selected using RFE-SVM(Linear). For all the classifiers used, this technique experimentally demonstrates an increase in performance metrics.

One of the most challenging problems of ultrasonic non-destructive testing is the signal distortion caused by the presence of noise, yielding the sound wave corruption and thus degrading the ultrasonic imaging technology performance due to Time of flight methods’ loss of precision. Deep learning algorithms have proven their effectiveness in reducing noise on several types of signals in different domains. In this paper, we propose a one-dimensional convolutional autoencoder for ultrasonic signal denoising. The efficiency of the proposed architecture is compared to the wavelet decomposition method, collating the peak signal-to-noise ratio values on the denoised signals. Our method proved its potential for NDT applications in recovering temporal information even on very noisy signals, and improving the PSNR by about 30 dB.

This paper reviews two types of writing recognition and details the phases incorporated on the online phase, which includes pretreatment, segmentation, characteristics extraction, recognition and decision.

In this paper, we propose to characterize the on-line handwriting for the early detection of Parkinson’s disease. Thus, using kinematics, mechanical, and spatial features of handwriting, we are looking for the characterization of Parkinson’s disease. This paper describes the phase of the data acquisition which is currently carried out with in the Neurological department of UHC Hassan II of Fez. Following this paper, we have proposed an approach based on unsupervised learning techniques for analyzing on-line handwriting of 34 Parkinson’s disease patients and 34 Healthy Controls according to quantitative and qualitative features. Based on 230 computed features for each participant, our study has uncovered three different types of writers. The results show that the complications of fine motor abilities in Parkinson’s disease patients is especially characterized by a significant degradation in handwriting kinematic features.

The current paper gives an overview on the technologies used in word spotting especially for handwritten documents. We present recent interesting published researches in this domain. We firstly give a general overview on the steps of the word spotting system, and then we describe a framework of techniques and methods used in each step, at the end we present a summary of these research in a table.

The researchers have focused on the techniques used to create systems for data protection such as signature verification, investigation in crimes, handwriting mails, graphical documents analysing, human assisting, etc. Recently, several researchers have focused on creating an optimal system for data processing in handwriting documents. This paper aims to present an overview about the studies and techniques used to achieve this system. The paper presents at first the most used databases, then a general pipeline diagram for word spotting system, next in each step (pre-processing, segmentation, feature extraction and classification) the most used techniques/methods and a summary table about the published researches are presented and finally the latest studies in this field.

Parkinson's disease (PD) is the second most common neurodegenerative disease affecting significant portion of elderly population. One of the most frequent hallmarks and usually also the first manifestation of PD is deterioration of handwriting. Since the diagnosis of Parkinson's disease is difficult, researchers have worked to develop a support tool based on algorithms to separate healthy controls from PD patients. Online handwriting analysis is one of the methods that can be used to diagnose PD.In this study, we aimed to analyze the Arabic Handwriting of 18 Parkinson's disease (PD) patients and 18 age matched healthy controls. We focused on copying an Arabic text task. For each participant we have calculated 528 features, and the purpose of this study is to find a subset of selected handwriting features suitable for efficiently identifying subjects with PD. The selected features were fed to a support vector machine classifier with RBF kernel, whose aim is to identify the subjects suffering from PD. Robustness of this classifier is measured with three performance matrices i.e., accuracy, sensitivity, and specificity. The obtained results show almost 80% overall classification accuracy.

The similarity or the distance measure have been used widely to calculate the similarity or dissimilarity between vector sequences, where the document images similarity is known as the domain that dealing with image information and both similarity/distance has been an important role for matching and pattern recognition. There are several types of similarity measure, we cover in this paper the survey of various distance measures used in the images matching and we explain the limitations associated with the existing distances. Then, we introduce the concept of the floating distance which describes the variation of the threshold’s selection for each word in decision making process, based on a combination of Linear Regression and cosine distance. Experiments are carried out on a handwritten Arabic image documents of Gallica library. These experiments show that the proposed floating distance outperforms the traditional distance in word spotting system.

Parkinson's disease (PD) is the second most common neurodegenerative disease affecting significant portion of elderly population. One of the most frequent hallmarks and the first manifestation of PD is deterioration of handwriting. Since the diagnosis of Parkinson's disease is difficult, researchers have worked to develop a support tool based on algorithms to separate healthy controls from PD patients. On-line handwriting analysis is one of the methods that can be used to diagnose PD. In this study, we aimed to analyze the Arabic Handwriting of 28 Parkinson's disease patients and 28 healthy controls (HCs) who were the same age and have the same intellectual level. We focused on copying an Arabic text task. For each participant we have calculated 1482 features. Based on the most relevant features selected by the Pearson's coefficient correlation, the Hierarchical Ascendant Classification (HAC) was applied and generated 3 clusters of writers. The characterization of these clusters was carried out by using the quantitative and qualitative parameters. The obtained results show that the combination of these two aspects can discriminate at best PD patients from HCs.

The old manuscripts are a part of the richest cultural heritage and legacy of civilizations where the digitalization is a solution for the preservation of these manuscripts. The conception of handwriting recognition system knows today a great expansion and appears as a necessity in order to exploit the wealth of information contained in ancient manuscripts. In this paper, a holistic approach for spotting and searching query, especially, for images documents in handwritten Arabic is proposed. These operations need a lot of time and effort to do manual work. For this, we use in the first time text line segmentation of handwritten document image based on partial projection, where a sliding-window approach is used to locate the document regions that are most similar to the query. Histograms of Oriented Gradients (HOGs) are used as the feature vectors to represent the query and documents image, then Support Vector Machines (SVM) is used to produce a better representation of the query and to classify feature vectors. Finally, the application of the reclassification technique at the indexation stage, leads to better results. General Terms Pattern Recognition.

International audience; L'objectif de cet article est d'analyser l'écriture manuscrite en ligne, caractérisée par une intervention primordiale des muscles et du cerveau, afin de détecter des pathologies neurologiques comme l'Alzheimer et le Parkinson, qui sont liées à l'organisation de l'ensemble des facultés mentales. Dans ce cadre, une acquisition de données est réalisée au sein du service neurologie au CHU Hassan II de Fès. Elle est faite sur une tablette graphique WACOM permettant de récupérer les données spatiotemporelles de l'écriture manuscrite. La problématique de la détection de pathologies nécessite préalablement l'étude des paramètres descriptifs de l'écriture manuscrite permettant la caractérisation des sujets contrôles et des personnes atteintes. Cet article décrit principalement la phase d'acquisition de données en cours au CHU de Fès, ainsi que l'analyse préliminaire de paramètres spatiotemporels de l'écriture réalisée sur ces données

International audience; Skew angle detection is one of the most important component of Optical Characters Recognition (OCR) systems and documents analysis. Taking documents by the blind and visually impaired people with a mobile phone always have a degree of inclination. In this paper, a novel and an efficient method based on extraction of Harris corner features points and Hough transform is presented to estimate skew angle of printed documents. A comparative study using the ICDAR2015 database with the wellknown methods in literature, shown the effectiveness and the performance of the proposed algorithm

This paper presents an efficient word spotting system applied to handwritten Arabic documents, where images are represented with bag-of-visual-SIFT descriptors and a sliding window approach is used to locate the regions that are most similar to the query by following the query-by-example paragon. First, a pre-processing step is used to produce a better representation of the most informative features. Secondly, a region-based framework is deployed to represent each local region by a bag-of-visual-SIFT descriptors. Afterward, some experiments are in order to demonstrate the codebook size influence on the efficiency of the system, by analyzing the curse of dimensionality curve. In the end, to measure the similarity score, a floating distance based on the descriptor’s number for each query is adopted. The experimental results prove the efficiency of the proposed processing steps in the word spotting system.

In this paper we present an unsupervised segmentation-free method for spotting and searching query, especially, for images documents in handwritten Arabic, for this, Histograms of Oriented Gradients (HOGs) are used as the feature vectors to represent the query and documents image. Then, we compress the descriptors with the product quantization method. Finally, a better representation of the query is obtained by using the Support Vector Machines (SVM).

Ground-penetrating radar (GPR) is a geophysical method used for subsurface targets identification. However, a sophisticated processing of GPR data is necessary to extract targets signature.

This paper attempts to deal with the problems of query-by-example word spotting in handwritten Arabic document. This operation needs a lot of time and effort to do manual work. For this, we propose a fully non supervised methodology dedicated to the word spotting system, without using any text word or line segmentation step, because any segmentation errors of the document affect the subsequent word representations and matching steps. This work is addressing by using a method that integrates two steps into one. First the interest points are automatically extracted from the images. Then, the matching of the extracted points is established with a correlation technique combined to a relaxation technique. Validation study is conducted under a series of controlled experiments on handwritten Arabic document images.

International audience; Old manuscripts are a part of the richest cultural heritage and legacy of civilizations. Repetitive manual manipulation of fragile documents should be avoided as it could destroy them. Digitalization, therefore, is a convenient solution for the preservation of these manuscripts. Many digitization projects, which treat Latin scripts, have been developed such as manuscripts d'Oc and d'Oïl in the Vatican Library (IRHT 2011), Better Access to Manuscripts and Browsing of Images (Calabretto et al. 1999), etc. The conception of recognition systems for degraded handwritten Arabic document images knows today a great expansion and appears as a necessity in order to exploit the wealth of information contained in ancient manuscripts