A workflow management system to feed digital libraries: proposal and case study

Building a digital library of antique documents involves not only technical implementation issues, but also aspects related to the digitization of large collections of documents. Antique documents are usually delicate and need to be handled with care. Also, a poor state of preservation and the use of unrecognizable font types make automatic text recognition more difficult, hence requiring a further human revision to perform text corrections. This makes the participation of experts in the digitization process mandatory and, therefore, costly. In this paper, we present a framework for managing the workflow of the digitization of large collections of antique documents. We describe the digitization process, and a tool supporting all of its phases and tasks. We also present a case study in which we describe how the workflow management system was applied to the digitization of more than 10,000 documents from journals of the 19th century. In addition, we describe the resulting digital library, focusing on the most important technological issues.     

最大差值图决策的低照度图像自适应增强算法

应用于光照分布不均的低照度图像,传统的图像增强算法会出现色彩失真、亮区过度增强等问题,因此提出一种最大差值图决策的低照度图像自适应增强算法。首先,提出最大差值图的概念,通过最大差值图粗略估计出初始光照分量;然后,提出交替引导滤波的算法,利用交替引导滤波对初始光照分量进行校正,实现光照分量的准确估计;最后,设计了图像亮度自适应的伽马变换,能够根据获取的光照分量自适应调整伽马变换参数,从而在增强图像的同时消除光照不均带来的影响。实验结果表明,增强后的图像有效消除了光照分布不均带来的影响,图像亮度、对比度、细节表现能力和色彩保真度都得到了明显提升,平均梯度提升了1倍以上,信息熵提升了14%以上。由于提出的算法对光照分量估计准确,自适应伽马变换针对低照度图像进行了优化,因此,对于夜间等弱光源条件下的彩色图像具有十分有效的增强效果。     

Restoring 2D content from distorted documents

This paper presents a framework to restore the 2D content printed on documents in the presence of geometric distortion and non-uniform illumination. Compared with textbased document imaging approaches that correct distortion to a level necessary to obtain sufficiently readable text or to facilitate optical character recognition (OCR), our work targets nontextual documents where the original printed content is desired. To achieve this goal, our framework acquires a 3D scan of the document's surface together with a high-resolution image. Conformal mapping is used to rectify geometric distortion by mapping the 3D surface back to a plane while minimizing angular distortion. This conformal "deskewing" assumes no parametric model of the document's surface and is suitable for arbitrary distortions. Illumination correction is performed by using the 3D shape to distinguish content gradient edges from illumination gradient edges in the high-resolution image. Integration is performed using only the content edges to obtain a reflectance image with significantly less illumination artifacts. This approach makes no assumptions about light sources and their positions. The results from the geometric and photometric correction are combined to produce the final output.     

基于梯度场均衡化的图像对比度增强

在偏微分方程理论框架下提出一种能够有效增强图像中阴影或高亮区域信息的方法.首先对图像梯度场进行直方图均衡化,使这些图像阴影或高亮区域中的细节能够在梯度域得到增强;然后利用最小二乘原理重建出增强后的结果图像.通过引入 Lab 彩色空间将对比度增强方法推广到对彩色图像的处理中.在数值求解方面,根据Laplacian 算子的特点改进了求解 Poisson 方程的快速算法,改进后的算法具有程序设计简单、计算量小的特点.实验结果表明,文中方法能够有效地改善由于光照影响造成的图像对比度下降.     

一种数字航空影像的匀光方法

针对单张数字航空影像不均匀光照现象的成因，深入研究了马斯克匀光技术在数字航空影像匀光中的应用，并针对数字航空影像提出了相应的匀光处理的具体流程和实现方法。实验表明该方法可以克服数学模型法的不足，具有较强的适用性，对于消除数字航空影像的不均匀光照现象能够取得满意的效果，从而可以有效地解决单张数字航空影像的色彩平衡问题。     

抑制光照不均匀影响的图像特征点提取算法

为解决在光照不均匀情况下图像特征点提取算法表现效果不佳的问题,提出了一种改进的尺度不变特征转换（Scale Invariant Feature Transform,SIFT）算法抑制光照不均的影响。该方法在尺度空间构造中对输入的图像进行频域上的高斯高通滤波处理来滤除光照成分,并结合Top-hat变换弱化高斯滤波器参数选取难度,利用高斯卷积构建基于光照滤除与参数弱化的高斯差分金字塔,融合SIFT算法生成具有良好光照不变性的GT-SIFT描述子,进行特征点提取与匹配。实验结果表明,与传统算法相比改进算法在光照不均匀条件下具有更好的鲁棒性,图像特征点提取与匹配效果更好。     

基于同态滤波与Curvelet变换的钻孔图像自适应增强

针对岩石数字钻孔图像存在的光照不均、图像中岩石表面边缘细节模糊等情况,提出了一种钻孔图像自适应增强算法.对原图进行同态滤波;使用Curvelet变换分解原图与滤波后的图像,对两者的低频子带使用系数直方图匹配算法,将前者与后者的直方图进行匹配,改善光照不均的影响;对原图的高频子带使用自适应的阈值进行滤波,同时利用自适应增强函数进行增强;使用Curvelet反变换重构得到增强后的图像.实验结果表明:算法可以有效地改善钻孔图像光照不均的问题,增强图像中物体的边缘信息,在主观视觉效果和图像客观评价指标上相对于其他算法均有一定优势.     

Shape from shading for the digitization of curved documents

Document digitization is faster and more affordable using digital cameras than scanners. On the other hand, if we aim at extending the basic digital camera functionalities for such a purpose, post-processings will be of first importance, at least to improve the text legibility. In this paper, we address the specific problem of the virtual flattening of curved documents, as for example the pages of an opened book lying on its spine. In order to compute the document shape, we use the shape from shading technique and discuss why, in some cases, it is more suitable than other 3D single-view reconstruction techniques. We extend the seminal work by Wada et al. (Proceedings of the IAPR Workshop on machine vision and applications, Tokyo, Japan, pp. 591–594, 1992) and consecutive papers, reformulating the problem in terms of perspective shape from shading. Finally, we design a complete post-processing algorithm and test it on real images. Even if the documents are much curved, it is shown that the restored images are almost identical to scanned images of the flattened documents.     

Detection and analysis of table of contents based on content association

As a special type of table understanding, the detection and analysis of tables of contents (TOCs) play an important role in the digitization of multi-page documents. Most previous TOC analysis methods only concentrate on the TOC itself without taking into account the other pages in the same document. Besides, they often require manual coding or at least machine learning of document-specific models. This paper introduces a new method to detect and analyze TOCs based on content association. It fully leverages the text information throughout the whole multi-page document and can be directly applied to a wide range of documents without the need to build or learn the models for individual documents. In addition, the associations of general text and page numbers are combined to make the TOC analysis more accurate. Natural language processing and layout analysis are integrated to improve the TOC functional tagging. The applications of the proposed method in a large-scale digital library project are also discussed.     

高斯尺度空间下估计背景的自适应阈值分割算法

为有效分割非均匀光照图像,提出一种在高斯尺度空间下估计背景的自适应阈值分割算法. 首先,利用二维高斯函数对待处理图像进行卷积操作来构建一个高斯尺度空间,在此空间下进行背景估计,并采用背景差法来消除非均匀光照干扰,从而提取出目标图像;然后,采用 矫正进行增强处理以突出较暗目标信息;最后,经强调谷底的最大类间方差法进行全局分割得到最终结果. 为验证算法的有效性,对非均匀光照条件下文本图像以及非文本图像进行了测试,并与基于偏移场的模糊C均值方法、灰度波动变换自适应阈值分割算法和自适应最小误差阈值分割算法,在错误分割率和运行时间上进行了对比. 实验结果表明,对比以上三种方法,该算法的分割结果更为理想.     

Watermarking of Electronic Text Documents

With the tremendous development of the Internet, it has become desirable to distribute text documents electronically. However, commercial publishers may be reluctant to offer valuable digital documents online for the fear that they will be re-transmitted or copied illegally. To address this problem, we propose a robust watermarking technique, whereby electronic text documents are fingerprinted with one or more semantics-preserving modifications to the document text. The text modifications may be selected so that multiple copies of the same master document will all have the same meaning. By examining text modifications in an unauthorized copy, one can identify the authorized source and the recepient. In this paper, we present a new method that is accurate, robust against attacks (e.g., the cyber pirate may post only a section or a paragraph of a registered text online), scalable (e.g., a few pages of text to hundreds of pages) and secure (e.g., remove or modify embedded watermark with or without knowledge of watermarking method). This approach could therefore facilitate e-commerce of newspapers, journals, magazines, and in general any electronic text document possessing commercial value.     

Building cameras for capturing documents

This paper explores those aspects of document capture that are specific to cameras. Each of them must be addressed in order to close the gap between taking a photograph of a document and capturing the document itself. We present results in five areas: (1) framing documents using structured light, (2) robustly dealing with ambient illumination when capturing glossy documents, (3) improving text quality when using mosaiced color sensors, (4) robustly and passively recovering perspective and image plane skew using text flow, and (5) measuring and undoing page curl using structured light and an applicable surface model. The ultimate success of subsequent document recognition will be heavily dependent on the successful completion of these tasks.Received: 8 December 2003, Revised: 6 April 2004, Published online: 11 March 2005     

一种光照不均文本图像的校正算法

针对光照不均的文本图像在二值化后文字识别率底下的问题,文章提出了一种针对光照不均图像进行二值化和图像增强的算法。针对图像进行分块处理,根据每块自身的平均灰度值和均方差以及全图的平均灰度值来进行算法的自适应变换,已达到对不同光照的区块进行不同处理的目的。实验结果表明本算法能够比较有效的完成对光照不均的文本图像的处理,比单纯的二值化处理方法有一定的提高。     

Face recognition under varying illumination based on a 2D face shape model

This paper proposes a novel illumination compensation algorithm, which can compensate for the uneven illuminations on human faces and reconstruct face images in normal lighting conditions. A simple yet effective local contrast enhancement method, namely block-based histogram equalization (BHE), is first proposed. The resulting image processed using BHE is then compared with the original face image processed using histogram equalization (HE) to estimate the category of its light source. In our scheme, we divide the light source for a human face into 65 categories. Based on the category identified, a corresponding lighting compensation model is used to reconstruct an image that will visually be under normal illumination. In order to eliminate the influence of uneven illumination while retaining the shape information about a human face, a 2D face shape model is used. Experimental results show that, with the use of principal component analysis for face recognition, the recognition rate can be improved by 53.3% to 62.6% when our proposed algorithm for lighting compensation is used.     

一种光照不变人脸识别的预处理算法

提出了一种新的光照不变人脸识别的图像预处理算法称为分段局部归一化方法(SLN)。其思想是对图像像素分段,使得每段中各像素对应的物体表面点具有相近的表面法向量分布,因而对光源具有相似的灰度响应,然后局部归一化在各段中进行以削弱光照影响。该算法首先建立物体的朗伯(Lambert)表面反射模型,用奇异值分解方法估计出人脸形状的平均表面法向量分布矩阵,根据法向量方向利用聚类算法对像素进行分段,然后在各段中进行局部的像素归一化处理,最后传统的人脸识别算法如PCA在归一化后的图像中进行。在Harvard和YaleB人脸图像库中的识别试验表明,该算法能有效地提高在非均匀光照条件下的人脸识别率。     

光照鲁棒的模糊C均值聚类显微图像分割

传统的模糊C均值FCM聚类图像分割算法在显微图像分割中由于没有考虑光照不均匀的影响而降低了分割的效果,为此,提出了一种光照鲁棒的FCM显微图像分割算法。该算法用正交基函数的线性组合模拟不均匀光照,并引入到FCM算法的目标函数中,进行图像的模糊分割。算法不仅降低了非均匀光照对分割效果的影响,还可以同步估计不均匀光照场。实验结果表明,该方法非常有效。     

小波变换在图像照度不均校正中的应用

数字图像中照度不均匀影响了图像分割的质量,本文研究图像照度不均匀的校正。讨论几种现有数字图像照度不均匀的校正技术,分析这些方法在计算误差上面的原因及缺点,在此基础上提出基于小波变换的数字图像照度不均匀校正技术。小波变换兼顾了空域和频域,对光照不均匀数字图像做校正,这对后期图像分割非常重要。图像对比实验表明,该方法与其他方法相比较,具有速度快,所得图像区分度高、背景噪声小的特点。     

Flaw detection of domed surfaces in LED packages by machine vision system

Epoxy-packaging is widely used in light-emitting diode (LED) packages to protect LED chips and magnify the chip light. Surface flaws in LED packages affect not only the appearances of LEDs but also their functionality, efficiency and stability. Due to the high demand for productivity and quality, bare-eye-inspection approach becomes extremely inadequately. Therefore, this research proposes a machine-vision-based system for detecting tiny flaws occurred in the domed surfaces of LED epoxy-packing. We apply grey relational analysis to the frequency components in block discrete cosine transform domain, and significantly attenuate the large-magnitude frequency components that represent the background texture of the surface based on their corresponding grey relational grades. Then, by reconstructing the declined frequency components, we eliminate not only random texture but also uneven illumination patterns and retain anomalies in the restored image. This approach overcomes the difficulties of inspecting tiny flaws from uneven illumination backgrounds. Experimental results show that the proposed method can effectively inspect tiny flaws in LED domed surfaces.     

自适应最小误差阈值分割算法

对二维最小误差法进行三维推广, 并结合三维直方图重建和降维思想提出了一种鲁 棒的最小误差阈值分割算法. 但该方法为全局算法, 仅适用于分割均匀光照图像. 为 提高其自适应性, 本文采用Water flow模型对非均匀光照图像进行背景估计, 以此获 得原始图像与背景图像的差值图像, 达到降低非均匀光照对图像分割造成干扰的目的. 为进 一步提高分割性能, 本文对差值图像采用γ 矫正进行增强, 然后采用鲁棒最小误差 法进行全局分割, 从而完成目标提取. 最后本文对均匀光照下以及非均匀光照下图像进行了 实验, 并与一维最小误差法、二维最小误差法、三维直方图重建和降维的Otsu阈值分割 算法、灰度波动变换自适应阈值方法以及一种改进的FCM方法在错误分割率和运行时间上进 行了对比. 实验结果表明, 相对于以上方法, 本算法的分割性能均有明显提升.