基于多尺度的蒙古文脱机手写识别方法

蒙古文的一大特点是字符无缝连接,因此一个蒙古文单词有多种字符划分方式。根据蒙古文这一特点,该文提出了多尺度蒙古文脱机手写识别方法,即让一个手写蒙古文单词图像对应多种目标序列,用多个目标序列同时约束训练模型,使得模型更加精准地学习手写图像的细节信息和蒙古文构词规则。该文提出了“十二字头”码、变形显现码和字素码3种字符划分方法,且拥有相互包含关系,即“十二字头”码可以分解为变形显现码、变形显现码可以进一步分解为字素码。多尺度模型首先用多层双向长短时记忆网络对序列化手写图像进行处理,之后加入第一层连接时序分类器做“十二字头”码序列的映射,然后是第二层连接时序分类器做变形显现码序列的映射,最后是第三层连接时序分类器做字素码序列的映射。用三个连接时序分类器损失函数的和作为模型的总损失函数。实验结果表明,该模型在公开的蒙古文脱机手写数据集MHW上表现出了最佳性能,在简单的最佳路径解码方式下,测试集Ⅰ上的单词识别准确率为66.22%、测试集Ⅱ上为63.97%。     

基于字素分割的蒙古文手写识别研究

隐马尔科夫模型(HMM)对序列数据有很强的建模能力,在语音和手写识别中都得到了广泛的应用。利用HMM研究蒙古文手写识别,首先需要解决的问题是手写文字的序列化。从蒙古文的构词和书写特点看,蒙古文由多个字素从上到下串联构成。选择字素集合和词的字素分割是手写识别的基础,也是影响识别效果的关键因素。该文根据蒙古文音节和编码知识确定了蒙古文字母集合,共包括1 171个字母。通过相关性处理、HMM排序筛选等手段得到长字素集合,共包括378个字素。对长字素经过人工分解,获得了50个短字素。最后利用两层映射给出了词转字素序列的算法。为了验证长短字素在手写识别中的效果,我们在HTK(hidden Markov model toolkit)环境下利用小规模字库实现了手写识别系统,实验结果表明短字素比长字素有更好的性能。文中给出的字素集合和词转字素序列的算法为后续基于HMM的蒙古文手写识别研究奠定了基础。     

基于统计翻译框架的蒙古文自动拼写校对方法

在以国际标准编码存储的传统蒙古文电子文本中,拼写错误十分普遍。人工校对这些错误不仅速度慢而且成本高。该文提出了一种基于统计翻译框架的传统蒙古文自动拼写校对方法,将拼写校对看作是从错误词到正确词的翻译。该文使用改进的基于短语的统计机器翻译模型来构建拼写校对模型,然后对测试文本进行校对。实验结果表明,该方法可以快速、有效地校对拼写错误,而且不依赖于特定语言的语法知识。使用该方法对包含1 026个正确词、1 102个错误词的测试集进行拼写校对,校对后文本中的正确词所占比例最高可达97.55%。     

一种基于HMM和统计语言模型的维吾尔文及阿拉伯文识别方法

维吾尔文和阿拉伯文是采用阿拉伯文字母的从右向左书写的连写文字。它们识别方法的研究对于多文种文本图像内容的利用具有重要意义。利用HTK工具包,分别建立基于隐马尔科夫模型HMM(Hidden Markov Model)的印刷体维吾尔文和阿拉伯文识别系统,其中特征提取部分采用分布密度特征和局部方向特征。研究利用HTK工具建立维吾尔文和阿拉伯文统计语言模型,并将语言模型用于改进识别系统性能。实验结果表明采用统计语言模型可有效提高文字识别系统性能。其中,在包含24 000个单词的印刷体维吾尔文测试集上,通过利用语言模型识别率从78.28%提高到97.45%;在包含759个单词的印刷体阿拉伯文测试集上,通过利用语言模型识别率从79.07%提高到85.80%。     

基于深度卷积自编码神经网络的手写数字识别研究

针对提高不同笔体下的手写识别准确率进行了研究,将深度卷积神经网络与自动编码器相结合,设计卷积自编码器网络层数,形成深度卷积自编码神经网络。首先采用双线性插值方法分别对MNIST数据集与一万幅自制中国大学生手写数字图片进行图像预处理,然后先使用单一MNIST数据集对深度卷积自编码神经网络进行训练与测试;最后使用MNIST与自制数据集中5 000幅混合,再次训练该网络,对另外5 000幅进行测试。实验数据表明,所提深度卷积自编码神经网络在MNIST测试集正确率达到99.37%,有效提高了准确率;且5 000幅自制数据集模型测试正确率达99.33%,表明该算法实用性较强,在不同笔体数字上得到了较高的识别准确率,模型准确有效。     

基于分割识别的蒙古语语音关键词检测方法的研究

蒙古文属于黏着语,词根和后缀能够组合成近百万的蒙古文单词.现有的蒙古语大词汇量连续语音识别(LVCSR)系统的发音词典无法包含所有蒙古文单词.同时发音词典较大时,训练语料的稀疏将导致LVCSR系统的性能明显下降.为了解决LVCSR系统中大多数蒙古文单词的识别问题和蒙古语语音关键词检测系统中大量集外词的检测问题,结合蒙古文的构词特点,提出了基于分割识别的蒙古语LVCSR方法,并建立了对应的声学模型和语言模型.最后,将此方法应用到了蒙古语语音关键词检测系统中并在蒙古语语音语料上进行了测试.实验结果表明,基于分割识别的蒙古语LVCSR方法能解决大部分蒙古文单词的识别问题,并将蒙古语语音关键词检测系统的大量集外词转化成了集内词,大幅度提高了检测系统的查准率和召回率.     

基于部件组合的联机手写“藏文—梵文”样本生成

“藏文—梵文”包括500多个现代藏文、6 000多个梵音藏文,在文字识别领域属于大类别的字符集,所以联机手写样本采集是庞大而复杂的工程。鉴于此,提供了一种基于部件组合的“藏文—梵文”手写样本生成方法,主要包括: (1)确定“藏文—梵文”字符集和部件集;(2)获取“藏文—梵文”字丁的部件位置信息;(3)采集联机手写“藏文—梵文”部件的样本;(4)生成联机手写“藏文—梵文”字符集样本库。该文为联机手写“藏文—梵文”识别的研究提供字符训练样本库和测试样本库,提高了手写梵音藏文样本采集效率,解决了样本数量及多样性问题,降低了样本采集成本,为进一步联机手写梵音藏文识别的研究与系统开发奠定了基础。     

基于BERT与柱搜索的中文释义生成

释义生成任务是指为一个目标词生成相应的释义。该文在中文释义生成任务中使用了目标词的上下文信息,并提出了一个基于BERT与柱搜索的释义生成模型。该文构建了包含上下文的CWN中文数据集,同时也在Oxford英文数据集上开展了实验。实验结果显示,该文模型在中英文数据集上性能均有显著提升,其中CWN数据集实验结果相比基线模型BLEU指标提升了10.47,语义相似度指标提升了0.105。语义相似度指标与人工评价结果相关性更高。最后,该文分析了中文释义生成任务仍存在的四个问题。     

手写中文地址识别后处理方法的研究

OCR(光学字符识别技术)作为方便有效的字体识别技术,在办公自动化、信息恢复、数字图书馆等方面发挥着日益重要的作用。语言模型在OCR后处理,特别是在中文的文字识别后处理方面有着广泛的应用。本文针对手写中文地址的后处理,讨论了语言模型的粒度对识别正确率的影响,分析了基于字和基于词的语言模型各自的优点和缺点,并采用了基于词的语言模型,在此基础上提出了加权词图搜索算法。实验证明,在58269条中文手写地址的测试集上,手写地址的整体识别率由原来的28.56%上升到了75.66% ,错误率下降了65.93% ,大大提高了系统的性能。     

蒙古文字母到音素转换方法的研究

针对蒙古文字母到音素的转换(grapheme to phoneme conversion, G2P)问题, 提出了基于规则的蒙古文G2P转换方法和基于联合序列模型的蒙古文G2P转换方法。实验结果表明, 利用联合序列模型的蒙古文G2P转换方法要明显好于基于规则的蒙古文G2P转换方法。并且建立的基于联合序列模型的蒙古文G2P转换系统的词误识率为1632%, 音素误识率仅为337%, 能达到实用要求。     

Recognition-directed recovering of temporal information from handwriting images

This paper analyses a handwriting recognition system for offline cursive words based on HMMs. It compares two approaches for transforming offline handwriting available as two-dimensional images into one-dimensional input signals that can be processed by HMMs. In the first approach, a left–right scan of the word is performed resulting in a sequence of feature vectors. In the second approach, a more subtle process attempts to recover the temporal order of the strokes that form words as they were written. This is accomplished by a graph model that generates a set of paths, each path being a possible temporal order of the handwriting. The recognition process then selects the most likely temporal stroke order based on knowledge that has been acquired from a large set of handwriting samples for which the temporal information was available. We show experimentally that such an offline recognition system using the recovered temporal order can achieve recognition performances that are much better than those obtained with the simple left–right order, and that come close to those of an online recognition system. We have been able to assess the ordering quality of handwriting when comparing true ordering and recovered one, and we also analyze the situations where offline and online information differ and what the consequences are on the recognition performances. For these evaluations, we have used about 30,000 words from the IRONOFF database that features both the online signal and offline signal for each word.     

Handwritten Chinese text recognition by integrating multiple contexts

This paper presents an effective approach for the offline recognition of unconstrained handwritten Chinese texts. Under the general integrated segmentation-and-recognition framework with character oversegmentation, we investigate three important issues: candidate path evaluation, path search, and parameter estimation. For path evaluation, we combine multiple contexts (character recognition scores, geometric and linguistic contexts) from the Bayesian decision view, and convert the classifier outputs to posterior probabilities via confidence transformation. In path search, we use a refined beam search algorithm to improve the search efficiency and, meanwhile, use a candidate character augmentation strategy to improve the recognition accuracy. The combining weights of the path evaluation function are optimized by supervised learning using a Maximum Character Accuracy criterion. We evaluated the recognition performance on a Chinese handwriting database CASIA-HWDB, which contains nearly four million character samples of 7,356 classes and 5,091 pages of unconstrained handwritten texts. The experimental results show that confidence transformation and combining multiple contexts improve the text line recognition performance significantly. On a test set of 1,015 handwritten pages, the proposed approach achieved character-level accurate rate of 90.75 percent and correct rate of 91.39 percent, which are superior by far to the best results reported in the literature.     

Supporting electronic ink databases

The emergence of the pen as the main interface device for personal digital assistants and pen-computers has made handwritten text, and more generally ink, a first-class object. As for any other type of data, the need of retrieval is a prevailing one. Retrieval of handwritten text is more difficult than that of conventional data since it is necessary to identify a handwritten word given slightly different variations in its shape. The current way of addressing this is by using handwriting recognition, which is prone to errors and limits the expressiveness of ink. Alternatively, one can retrieve from the database handwritten words that are similar to a query handwritten word using techniques borrowed from pattern and speech recognition. In this paper, an indexing technique based on Hidden Markov Models is proposed. Its implementation and its performance is reported in this paper.     

基于加权贝叶斯的脱机手写阿文单词识别

针对手写阿拉伯单词书写连笔,且相似词较多的特点,该文提出一种新的脱机手写文字识别算法。该算法以固定组件为成分拆分阿拉伯单词,构建自组件特征至单词类别的加权贝叶斯推理模型。算法结合单词组件分割、多级混合式组件识别、组件加权系数估计等,计算单词类别的后验概率并得到单词识别结果。在IFN/ENIT库上的实验,获得了90.03%的单词识别率,证实组件分解对笔画连写具有鲁棒性,组件识别能提高相似词的辨别能力,而且该算法所需训练类别少,易向大词汇量识别扩展。     

SCUT-COUCH2009—a comprehensive online unconstrained Chinese handwriting database and benchmark evaluation

A comprehensive online unconstrained Chinese handwriting dataset, SCUT-COUCH2009, is introduced in this paper. As a revision of SCUT-COUCH2008 [1], the SCUT-COUCH2009 database consists of more datasets with larger vocabularies and more writers. The database is built to facilitate the research of unconstrained online Chinese handwriting recognition. It is comprehensive in the sense that it consists of 11 datasets of different vocabularies, named GB1, GB2, TradGB1, Big5, Pinyin, Letters, Digit, Symbol, Word8888, Word17366 and Word44208. In particular, the SCUT-COUCH2009 database contains handwritten samples of 6,763 single Chinese characters in the GB2312-80 standard, 5,401 traditional Chinese characters of the Big5 standard, 1,384 traditional Chinese characters corresponding to the level 1 characters of the GB2312-80 standard, 8,888 frequently used Chinese words, 17,366 daily-used Chinese words, 44,208 complete words from the Fourth Edition of “The Contemporary Chinese Dictionary”, 2,010 Pinyin and 184 daily-used symbols. The samples were collected using PDAs (Personal Digit Assistant) and smart phones with touch screens and were contributed by more than 190 persons. The total number of character samples is over 3.6 million. The SCUT-COUCH2009 database is the first publicly available large vocabulary online Chinese handwriting database containing multi-type character/word samples. We report some evaluation results on the database using state-of-the-art recognizers for benchmarking.     

The IAM-database: an English sentence database for offline handwriting recognition

In this paper we describe a database that consists of handwritten English sentences. It is based on the Lancaster-Oslo/Bergen (LOB) corpus. This corpus is a collection of texts that comprise about one million word instances. The database includes 1,066 forms produced by approximately 400 different writers. A total of 82,227 word instances out of a vocabulary of 10,841 words occur in the collection. The database consists of full English sentences. It can serve as a basis for a variety of handwriting recognition tasks. However, it is expected that the database would be particularly useful for recognition tasks where linguistic knowledge beyond the lexicon level is used, because this knowledge can be automatically derived from the underlying corpus. The database also includes a few image-processing procedures for extracting the handwritten text from the forms and the segmentation of the text into lines and words. Received September 28, 2001 / Revised October 10, 2001