Efficient use of training data in the n-tuple recognition method

A simple technique is presented for improving the robustness of the n-tuple recognition method against inauspicious choices of architectural parameters, guarding against the saturation problem, and improving the utilisation of small data sets. Experiments are reported which confirm that the method significantly improves performance and reduces saturation in character recognition problems.

Digitisation Processing and Recognition of Old Greek Manuscipts (the D-SCRIBE Project)

After many years of scholar study, manuscript collections continue to be an important source of novel information for scholars, concerning both the history of earlier times as well as the development of cultural documentation over the centuries. D-SCRIBE project aims to support and facilitate current and future efforts in manuscript digitization and processing. It strives toward the creation of a comprehensive software product, which can assist the content holders in turning an archive of manuscripts into a digital collection using automated methods. In this paper, we focus on the problem of recognizing early Christian Greek manuscripts. We propose a novel digital image binarization scheme for low quality historical documents allowing further content exploitation in an efficient way. Based on the existence of closed cavity regions in the majority of characters and character ligatures in these scripts, we propose a novel, segmentation-free, fast and efficient technique that assists the recognition procedure by tracing and recognizing the most frequently appearing characters or character ligatures.

Neural network based off-line handwritten text recognition system

This dissertation introduces a new system for handwritten text recognition based on an improved neural network design. Most of the existing neural networks treat mean square error function as the standard error function. The system as proposed in this dissertation utilizes the mean quartic error function, where the third and fourth derivatives are non-zero. Consequently, many improvements on the training methods were achieved. The training results are carefully assessed before and after the update. To evaluate the performance of a training system, there are three essential factors to be considered, and they are from high to low importance priority: (1) error rate on testing set, (2) processing time needed to recognize a segmented character and (3) the total training time and subsequently the total testing time. It is observed that bounded training methods accelerate the training process, while semi-third order training methods, next-minimal training methods, and preprocessing operations reduce the error rate on the testing set. Empirical observations suggest that two combinations of training methods are needed for different case character recognition. Since character segmentation is required for word and sentence recognition, this dissertation provides also an effective rule-based segmentation method, which is different from the conventional adaptive segmentation methods. Dictionary-based correction is utilized to correct mistakes resulting from the recognition and segmentation phases. The integration of the segmentation methods with the handwritten character recognition algorithm yielded an accuracy of 92% for lower case characters and 97% for upper case characters. In the testing phase, the database consists of 20,000 handwritten characters, with 10,000 for each case. The testing phase on the recognition 10,000 handwritten characters required 8.5 seconds in processing time.

An evaluation of the neocognitron

We describe a sequence of experiments investigating the strengths and limitations of Fukushima's neocognitron as a handwritten digit classifier. Using the results of these experiments as a foundation, we propose and evaluate improvements to Fukushima's original network in an effort to obtain higher recognition performance. The neocognitron's performance is shown to be strongly dependent on the choice of selectivity parameters and we present two methods to adjust these variables. Performance of the network under the more effective of the two new selectivity adjustment techniques suggests that the network fails to exploit the features that distinguish different classes of input data. To avoid this shortcoming, the network's final layer cells were replaced by a nonlinear classifier (a multilayer perceptron) to create a hybrid architecture. Tests of Fukushima's original system and the novel systems proposed in this paper suggest that it may be difficult for the neocognitron to achieve the performance of existing digit classifiers due to its reliance upon the supervisor's choice of selectivity parameters and training data. These findings pertain to Fukushima's implementation of the system and should not be seen as diminishing the practical significance of the concept of hierarchical feature extraction embodied in the neocognitron. © 1997 IEEE.

Kernels for large margin time-series classification

In this paper we propose a novel family of kernels for multivariate time-series classification problems. Each time-series is approximated by a linear combination of piecewise polynomial functions in a Reproducing Kernel Hilbert Space by a novel kernel interpolation technique. Using the associated kernel function a large margin classification formulation is proposed which can discriminate between two classes. The formulation leads to kernels, between two multivariate time-series, which can be efficiently computed. The kernels have been successfully applied to writer independent handwritten character recognition.

Resolving Ambiguities in Confused Online Tamil Characters with Post Processing Algorithms

This paper addresses the problem of resolving ambiguities in frequently confused online Tamil character pairs by employing script specific algorithms as a post classification step. Robust structural cues and temporal information of the preprocessed character are extensively utilized in the design of these algorithms. The methods are quite robust in automatically extracting the discriminative sub-strokes of confused characters for further analysis. Experimental validation on the IWFHR Database indicates error rates of less than 3 % for the confused characters. Thus, these post processing steps have a good potential to improve the performance of online Tamil handwritten character recognition.

Font and Size Independent OCR for Printed Kannada Documents using SVM Classifiers

The following topics were dealt with: document analysis and recognition; multimedia document processing; character recognition; document image processing; cheque processing; form processing; music processing; document segmentation; electronic documents; character classification; handwritten character recognition; information retrieval; postal automation; font recognition; Indian language OCR; handwriting recognition; performance evaluation; graphics recognition; oriental character recognition; and word recognition

Lexicon-free, novel segmentation of online handwritten indic words

Research in the field of recognizing unlimited vocabulary, online handwritten Indic words is still in its infancy. Most of the focus so far has been in the area of isolated character recognition. In the context of lexicon-free recognition of words, one of the primary issues to be addressed is that of segmentation. As a preliminary attempt, this paper proposes a novel script-independent, lexicon-free method for segmenting online handwritten words to their constituent symbols. Feedback strategies, inspired from neuroscience studies, are proposed for improving the segmentation. The segmentation strategy has been tested on an exhaustive set of 10000 Tamil words collected from a large number of writers. The results show that better segmentation improves the overall recognition performance of the handwriting system.

脉宽压缩光栅用的多层膜设计和性能分析

应用于啁啾脉冲放大技术中的脉宽压缩光栅是基于多层膜作为基底，利用全息干涉技术和离子束技术刻蚀而成。脉宽压缩光栅的衍射效率和抗激光损伤阈值一方面依赖于光栅结构的设计，另一方面很大程度上取决于作为基底的多层膜的设计。给出了以413．1nm作为写入波长，1053nm作为使用波长的多层介质光栅膜的设计．样品在ZZS-800F、型真空镀膜机上采用电子束蒸发方式沉积而成，并给出了膜系结构对光学性能影响因素的详细分析，结果表明膜系H3L(H2L)^9H0．5L2．03H满足光栅膜的指标。给出了样品光学特性测试，其使用波

基于数学形态学的联机手写字符识别去噪方法

手写输入时由于笔尖抖动等原因产生了大量噪声,有效地去除噪声是手写识别的前提和关键。根据联机手写识别中手写体字符形态的特性,分析了手写时由于各种原因而产生的噪声,运用数学形态学中膨胀、腐蚀、细化等基本运算,提出了一种将数学形态学应用于联机手写识别预处理的方法,该方法可以有效地消除大量的冗余信息。测试结果表明,提出的方法可行,具有很好的鲁棒性,可以配合其他方案应用于各种联机手写字符识别中。

混合模式识别系统研究

讨论基于多种分类方法的模块组合实现的混合模式识别系统，它不同于利用多分类器输出结果表决的集成系统．提出两个系统：一个面向印刷体汉字文本识别，另一个面向自由手写体数字识别．利用多种特征和多种分类方法的组合、部分识别信息控制混淆字判别策略以及提出的动态模板库匹配后处理方法，使系统的性能与传统单一分类器系统比较，获得明显改善．实验表明：多方法多策略混合是解决复杂和增强系统鲁棒性的一条途径

基于DM642的车牌识别系统设计

本文设计与实现了一种基于TMS320DM642的车牌识别系统,详细阐述了该系统的硬件构成、软件流程、检测算法以及针对DSP处理器进行的系统优化。系统通过摄像头获取汽车牌照图像,以TMS320DM642处理器为核心建立硬件平台,完成车牌定位,倾斜角校正,字符分割,字符识别等一系列算法。实验结果表明基于TMS320DM642的车牌识别系统准确、有效,应用前景广泛。

影响汉语阅读能力发展的相关认知因素及其作用的发展变化

The present cross-sectional study paid attention to Chinese reading acquisition of 391 children from preschool to grade 3 in two elementary schools, and investigated the relationship between orthographic processing skills, morphological awareness, phonological awareness, naming, phonological memory, visual processing skill and reading skills, after controlling the variance of age, nonverbal intelligence and pinyin knowledge. The main results are as follows: Firstly, there are many different language skills as the predictors of Chinese reading success. Orthographic processing skills, morphological awareness, phonological awareness and naming are important in single-character recognition and comprehension. Beside them, the effect of visual processing skill and phonological memory for comprehension are also significant. Among them, the role of orthographic processing skills is the most important, whatever in single-character recognition or in comprehension. Secondly, orthographic processing skills are the most important factors in reading acquisition at low grade and its effect drops obviously after grade 2. Thirdly, morphological awareness is also the factor that cannot be ignored whatever for single-character recognition or for comprehension. Its influence appears in preschool and becomes the only significant predictor of character recognition in grade 3. Furthermore, morphological awareness is more relevant with the development of comprehension. Fourthly, phonological awareness plays the secondary role in Chinese reading acquisition except in grade 2 when its contribution is most of all. And compare with morphological awareness, the effect of phonological awareness is relative low. Fifthly, naming is important through preschool to grade 2. The contribution of phonological memory increases from preschool to grade 3 in comprehension.