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.     

Web文本中维吾尔语领域术语的自动发现

针对维吾尔语领域术语获取难度大,人工扩充领域术语工作量大、效率低等特点,利用词汇共现原理,以维吾尔语连接词和互信息(MI)为工具,快速扩充原始维吾尔语领域术语;建立了以维吾尔语领域术语为特征模板,利用条件随机场(CRF)模型实现Web文本中维吾尔语领域术语的自动发现方法,并在此基础上实现长维吾尔语领域术语的自动发现。实验表明,对短维吾尔语领域术语的自动发现准确率为 97.59%,召回率为 93.38%,对长维吾尔语领域术语的自动发现正确率达到55.72%。     

Machine printed text and handwriting identification in noisy document images

In this paper, we address the problem of the identification of text in noisy document images. We are especially focused on segmenting and identifying between handwriting and machine printed text because: 1) Handwriting in a document often indicates corrections, additions, or other supplemental information that should be treated differently from the main content and 2) the segmentation and recognition techniques requested for machine printed and handwritten text are significantly different. A novel aspect of our approach is that we treat noise as a separate class and model noise based on selected features. Trained Fisher classifiers are used to identify machine printed text and handwriting from noise and we further exploit context to refine the classification. A Markov Random Field-based (MRF) approach is used to model the geometrical structure of the printed text, handwriting, and noise to rectify misclassifications. Experimental results show that our approach is robust and can significantly improve page segmentation in noisy document collections.     

Unsupervised writer adaptation applied to handwritten text recognition

This paper deals with the problem of off-line handwritten text recognition. It presents a system of text recognition that exploits an original principle of adaptation to the handwriting to be recognized. The adaptation principle is based on the automatic learning, during the recognition, of the graphical characteristics of the handwriting. This on-line adaptation of the recognition system relies on the iteration of two steps: a word recognition step that allows to label the writer's representations (allographs) on the whole text and a re-evaluation step of character models. Tests carried out on a sample of 15 writers, all unknown by the system, show the interest of the proposed adaptation scheme since we obtain during iterations an improvement of recognition rates both at the letter and the word levels.     

Automatic writer identification using connected-component contours and edge-based features of uppercase Western script

In this paper, a new technique for offline writer identification is presented, using connected-component contours (COCOCOs or CO3s) in uppercase handwritten samples. In our model, the writer is considered to be characterized by a stochastic pattern generator, producing a family of connected components for the uppercase character set. Using a codebook of CO3s from an independent training set of 100 writers, the probability-density function (PDF) of CO3s was computed for an independent test set containing 150 unseen writers. Results revealed a high-sensitivity of the CO3 PDF for identifying individual writers on the basis of a single sentence of uppercase characters. The proposed automatic approach bridges the gap between image-statistics approaches on one end and manually measured allograph features of individual characters on the other end. Combining the CO3 PDF with an independent edge-based orientation and curvature PDF yielded very high correct identification rates.     

A novel method for binarization of scene text images and its application in text identification

Pattern Analysis and Applications - The aim of this article is twofold. First, we propose an effective methodology for binarization of scene images. For our present study, we use the publicly...     

Multi-label personality trait identification from text

Understanding the personality is beneficial for many purposes, e.g., it is natural to predict a user’s personality before offering him or her any services. The personality is intrinsic in the behavior of a person in all aspects, such as text writing. Some work has been proposed in recent times for correctly classifying a person’s personality from the text. However, it is still a significant challenge as the achieved accuracy is low; therefore, the proposed work addresses this issue. Effective feature selection techniques provide better classification accuracy in multi-label classification and personality traits identification as multi-label classification problem requires efficacy of feature selection methods. Therefore, to improve the accuracy using feature selection technique, this paper proposes a method for personality trait recognition from textual data called P ersonality T rait Classification based on L inguistic and F eature selection as M ulti-label classification (PTLFM). It combines analysis of variance’s F-statistic, Chi-square, and Mutual information with the sequential feature selection wrapper method to rank features. These three criteria apprehend different aspects of the dataset. The experimental results demonstrate that the proposed PTLFM method achieves higher accuracy across all the personality traits than the prevailing state-of-the-art machine learning and deep learning models. PTLFM provides an impressive absolute improvement of 2.23% and 3.84% of comparative improvement over the existing prevalent method, with more than 90% of features discarded. Furthemore, the proposed PTLFM achieves a percentage gain compared to the competitive methods across different personality traits Extraversion, Neuroticism, Agreeableness, Conscientiousness, and Openness in absolute terms 1.17, 1.94, 2.35, 1.64, and 0.35 respectively, and in comparative terms 2.01, 3.27, 4.14, 2.86, and 0.56 respectively. The results suggest that although deep learning is a popular paradigm, it does not always lead to a better predictive performance than machine learning models in all the problem domains.

     

Automatic expert system for weeds/crops identification in images from maize fields

Automation for the identification of plants, based on imaging sensors, in agricultural crops represents an important challenge. In maize fields, site-specific treatments, with chemical products or mechanical manipulations, can be applied for weeds elimination. This requires the identification of weeds and crop plants. Sometimes these plants appear impregnated by materials coming from the soil (particularly clays). This appears when the field is irrigated or after rain, particularly when the water falls with some force. This makes traditional approaches based on images greenness identification fail under such situations. Indeed, most pixels belonging to plants, but impregnated, are misidentified as soil pixels because they have lost their natural greenness. This loss of greenness also occurs after treatment when weeds have begun the process of death. To correctly identify all plants, independently of the loss of greenness, we design an automatic expert system based on image segmentation procedures. The performance of this method is verified favorably.     

Automatic identification of oceanic eddies in infrared satellite images

Oceanic eddies have a large impact on climate and human activities; consequently, it is worthwhile to characterise them. One of their main features is size; however, it is a difficult task to obtain user-independent estimates of this feature from brightness temperature maps for eddies near the Iberian Peninsula. The reason is that the current methods in the scientific literature are unable to handle the variability in the shape and size of these eddies as well as the weak temperature gradients associated with them, especially those found off Iberia or those methods employ user-defined values that influence the estimate of the eddies' sizes. Our new method solves these problems using orientation fields and clustering methods. Its outcome is an ellipse that characterizes the size of the eddies with good precision.     

文本环绕图像

巧妙运用Text Wrap（文本环绕）可以强化版面并吸引读者的关注。ImagineFX美术编辑PaulTysall如此解释说。     

从文本自动构建OWL本体的研究

随着语义检索的发展,近年来涌现了许多基于本体的研究和应用,但本体本身仍离不开领域专家手工或半自动化的构建,成为了本体研究领域的一个瓶颈。因此,本文着眼于本体的自动化构建,提出了一种用FCA(形式概念分析)从文本中提取并自动生成符合W3C标准的OWL通用本体库的方法。解决了目前本体构建自动化程度低,领域依赖性强的问题,使得本体的发展和应用不再是空中楼阁。     

Automatic eyeglasses removal from face images

In this paper, we present an intelligent image editing and face synthesis system that automatically removes eyeglasses from an input frontal face image. Although conventional image editing tools can be used to remove eyeglasses by pixel-level editing, filling in the deleted eyeglasses region with the right content is a difficult problem. Our approach works at the object level where the eyeglasses are automatically located, removed as one piece, and the void region filled. Our system consists of three parts: eyeglasses detection, eyeglasses localization, and eyeglasses removal. First, an eye region detector, trained offline, is used to approximately locate the region of eyes, thus the region of eyeglasses. A Markov-chain Monte Carlo method is then used to accurately locate key points on the eyeglasses frame by searching for the global optimum of the posterior. Subsequently, a novel sample-based approach is used to synthesize the face image without the eyeglasses. Specifically, we adopt a statistical analysis and synthesis approach to learn the mapping between pairs of face images with and without eyeglasses from a database. Extensive experiments demonstrate that our system effectively removes eyeglasses.     

Automatic identification of commodity label images using lightweight attention network

Neural Computing and Applications - Recent research has raised interest in applying image classification techniques to automatically identify the commodity label images for the business automation...