A hybrid method for robust car plate character recognition

Image-based car plate recognition is an indispensable part of an intelligent traffic system. The quality of the images taken for car plates, especially for Chinese car plates, however, may sometimes be very poor, due to the operating conditions and distortion because of poor photographical environments. Furthermore, there exist some “similar” characters, such as “8” and “B”, “7” and “T” and so on. They are less distinguishable because of noises and/or distortions. To achieve robust and high recognition performance, in this paper, a two-stage hybrid recognition system combining statistical and structural recognition methods is proposed. Car plate images are skew corrected and normalized before recognition. In the first stage, four statistical sub-classifiers recognize the input character independently, and the recognition results are combined using the Bayes method. If the output of the first stage contains characters that belong to prescribed sets of similarity characters, structure recognition method is used to further classify these character images: they are preprocessed once more, structure features are obtained from them and these structure features are fed into a decision tree classifier. Finally, genetic algorithm is employed to achieve optimum system parameters. Experiments show that our recognition system is very efficient and robust. As part of an intelligent traffic system, the system has been in successful commercial use.     

Automated entry system for Chinese printed documents

In this paper, we present a new automated Chinese printed document entry system. This system features automated text/ graph segmentation, and multi-font, multi-size printed Chinese character recognition. Experimental results show that 95.8–99.4% of the top 10 printed characters can be correctly recognized, with the speed of 0.16 seconds/character.     

一个印刷体汉字识别系统的设计

给出一个印刷全汉字识别系统的设计方案，它主要包括扫描输入，模糊增强与聚类分割 ，图象数据二值比，通过并行神经网络进行汉字匹配等四个步骤。     

基于语义的手写汉字识别模型

目前,已经有很多文献阐述了不同的手写汉字识别算法,但是绝大多数算法都是针对单个汉字进行识别的,所以对于比较容易混淆的字,它们的识别效果都不好。该文针对这个问题,在单个汉字识别的基础上,结合汉语字典,加入了对前后汉字的语义考虑,大大地提高了这些容易混淆的汉字的识别率。     

Recognition of hand-printed Chinese characters using decision trees/machine learning C4.5 system

Recognition of Chinese characters has been an area of major interest for many years, and a large number of research papers and reports have already been published in this area. There are several major problems with Chinese character recognition: Chinese characters are distinct and ideographic, the character size is very large and a lot of structurally similar characters exist in the character set. Thus, classification criteria are difficult to generate. This paper presents a new technique for the recognition of hand-printed Chinese characters using the C4.5 machine learning system. Conventional methods have relied on hand-constructed dictionaries which are tedious to construct and difficult to make tolerant to variation in writing styles. The paper discusses Chinese character recognition using theHough transform for feature extraction and C4.5 system. The system was tested with 900 characters written by different writers from poor to acceptable quality (each character has 40 samples) and the rate of recognition obtained was 84%.     

基于小波变换的图像汉字笔划特征提取方法

借鉴仿生模式识别的认知观点,从汉字的构造机理和人类认识汉字的习惯角度出发,提出一种基于小波变换的图像汉字识别方法。制定了图像汉字笔划特征提取的具体规则,采用小波变换的方法对图像汉字边缘和笔划轮廓进行检测,通过有效提取图像汉字笔段信息,进行笔段合成,生成汉字或汉字的基本笔划。仿真实验结果表明,这种方法提高了图像汉字笔划特征提取的准确率和稳定性,对于印刷体和书写较规范的手写体图像汉字具有极高的识别率。     

多字体多字号印刷汉字识别方法的研究

本文对多体多字号印别汉字识别的方法进行了研究, 本文提出的方法是首先对不同字号印刷 汉字进行归一化处理, 再抽取汉字四周笔端数特征、改进粗外围特征、笔划穿插次数特征和投影变换特征, 然后对组合特征进行多级分类识别。实验在IBM一PC AT 微型机上进行, 结果表明, 实验系统在识别实际印别文本时识别率大于98%。     

基于组件合并的手写体汉字串分割

人们对孤立的手写体汉字字符的离线 识别做了大量的研究工作,而走向实用化的进展并不快.除了单字识别率不理想以外,从文本 中正确分割出单个汉字字符也是一个主要难题,因为字符的识别离不开正确分割.利用汉字的 基本结构特征,根据两个组件之间的上下、左右和包围关系,对组件进行合并形成完整的汉字 图像.对整个汉字字符串中组件的宽度和相邻组件的间距进行分析,有助于左右关系组件的合 并.实验结果表明,该方法对手写体汉字字符串具有理想的分割效果.     

Feature Point Method of Chinese Character Recognition and Its Application

A new method for recognizing Chinese characters is proposed. It is based on the so-called featurepoints of Chinese characters. The feature points we use include those on the stroke of a character, i.e., endpoints, turning points, fork points and cross points, and the key points on the background of character. Thismethod differs from the previous ones for it combines the feature points on stroke with those on back-ground and it uses feature points to recognize Chinese characters directly. A Chinese character recognitionsystem based on top-down dynamical matching of feature point is developed. The system can recognizenot only 6763 printed sample Song font Chinese characters of size 5.6×5.6mm~2 with high recognition rate,but also the general printed books, magazines and documents with a satisfactory recognition rate andspeed.     

On the almost-sure sample stability of systems with randomly time-varying delays

In a large class of multi-loop control systems, many feedback loops are “closed” through a time-shared digital computer, by means of algorithms which require information from sources which are sampled at a rate which is not synchronized with the sampling of the individual “plants”. This mis-synchronization, coupled with variations in the computer's task load caused by “interrupts”, results in a randomly time-varying delay in the closing of the various feedback loops. Consequently, the dynamics of each controlled “plant” in such a system may be modeled by means of a stochastic delay-differential equation. This paper presents some new research results concerning the sample stability, as opposed to statistical, or ensemble stability, of linear stochastic delay-differential equations.     

Hidden markov model based optical character recognition in the presence of deterministic transformations

A method is introduced to combine and jointly optimize recognition and image normalization in optical character recognition algorithms based on pseudo two-dimensional (2D) hidden Markov models (HMMs). The method can be combined with a previous method for joint segmentation and recognition of connected text. It also provides a maximum likelihood estimate of the transformation parameters (scaling factor, slant angle, etc.), that can be used by higher level modules in an intelligent document recognition system as an aid in the recognition process. The computational cost of this technique is modest. Experimental results on a data base of distorted printed characters are presented.     

Spring algorithms and symmetry

Spring algorithms are regarded as effective tools for visualizing undirected graphs. One major feature of applying spring algorithms is to display symmetric properties of graphs. This feature has been confirmed by numerous experiments. In this paper, firstly we formalize the concepts of graph symmetries in terms of “reflectional” and “rotational” automorphisms; and characterize the types of symmetries, which can be displayed simultaneously by a graph layout, in terms of “geometric” automorphism groups. We show that our formalization is complete. Secondly, we provide general theoretical evidence of why many spring algorithms can display graph symmetry. Finally, the strength of our general theorem is demonstrated from its application to several existing spring algorithms.     

Learning in analog neural network hardware

Hardware implementations of neuroprocessor architectures are currently enjoying commercial availability for the first time ever. This development has been caused in part by the requirement for real-time solutions to time critical neural network applications. Massively parallel asynchronous neuromorphic representations are inherently capable of very high computational speeds when properly cast in the “right stuff”, i.e. electronic or optoelectronic hardware. However, hardware based learning in such systems is still at a primitive stage. In practise, simulations are typically performed in software, and the resulting synaptic weight capturing the input-output transformation subsequently quantized and down-loaded onto the neural hardware. However, because of the numerous discrepancies between the software and hardware, such systems are inherently poor in performance. In this paper we report on chip-in-the-loop learning systems assembled from custom analog “building blocks” hardware.     

Artificial neural networks for document analysis and recognition

Artificial neural networks have been extensively applied to document analysis and recognition. Most efforts have been devoted to the recognition of isolated handwritten and printed characters with widely recognized successful results. However, many other document processing tasks, like preprocessing, layout analysis, character segmentation, word recognition, and signature verification, have been effectively faced with very promising results. This paper surveys the most significant problems in the area of offline document image processing, where connectionist-based approaches have been applied. Similarities and differences between approaches belonging to different categories are discussed. A particular emphasis is given on the crucial role of prior knowledge for the conception of both appropriate architectures and learning algorithms. Finally, the paper provides a critical analysts on the reviewed approaches and depicts the most promising research guidelines in the field. In particular, a second generation of connectionist-based models are foreseen which are based on appropriate graphical representations of the learning environment.     

Automatic Word Decompounding for ASR in a Morphologically Rich Language: Application to Amharic

This paper investigates a data-driven word decompounding algorithm for use in automatic speech recognition. An existing algorithm, called “Morfessor,” has been enhanced in order to address the problem of increased phonetic confusability arising from word decompounding by incorporating phonetic properties and some constraints on recognition units derived from forced alignments experiments. Speech recognition experiments have been carried out on a broadcast news task for the Amharic language to validate the approach. The out of vocabulary (OOV) word rates were reduced by 35% to 50% and a small reduction in word error rate (WER) has been achieved. The algorithm is relatively language independent and requires minimal adaptation to be applied to other languages.      

Unsupervised texture segmentation using Gabor filters

This paper presents a texture segmentation algorithm inspired by the multi-channel filtering theory for visual information processing in the early stages of human visual system. The channels are characterized by a bank of Gabor filters that nearly uniformly covers the spatial-frequency domain, and a systematic filter selection scheme is proposed, which is based on reconstruction of the input image from the filtered images. Texture features are obtained by subjecting each (selected) filtered image to a nonlinear transformation and computing a measure of “energy” in a window around each pixel. A square-error clustering algorithm is then used to integrate the feature images and produce a segmentation. A simple procedure to incorporate spatial information in the clustering process is proposed. A relative index is used to estimate the “true” number of texture categories.