Describing, transcribing, encoding, and editing modern correspondence material: a textbase approach

While letters and correspondence materials serve as (in)valuablesources of information for historians, philologists, (socio-)linguists,biographers, and textual critics, modern editorial theory merelyassigns them a secondary role. Contrary to this traditionaldocumentary view, the authors of this article argue for a treatmentof epistolary materials as primary sources in their own right.They propose a generalized text-base approach of encoded andannotated correspondence materials that can accomodate the generationof versatile user-driven electronic editions. This approachneeds to address current lacunae in markup theory and practice,resulting in a lack for either provisions for the encoding ofletter-specific phenomena in texts, or encoding features forsuch generative editions. A closer look at broader editorialtheories reveals a deeper lack of understanding of the natureand hence definition of correspondence materials. The authorspropose a Jakobsonian communicative definition of letters thatto a great deal can be mapped onto the textual model of theText Encoding Initiative (TEI). The second part of this articlediscusses the motivation for and practical realization of DigitalArchive of Letters in Flanders (DALF), a formal framework forencoding correspondence materials which is defined as a TEIcustomization. Its most important features for capturing detailedmetadata as well as letter-specific source phenomena are analysedand discussed against the text-ontological background sketchedout before.     

A multi-plane approach for text segmentation of complex document images

This study presents a new method, namely the multi-plane segmentation approach, for segmenting and extracting textual objects from various real-life complex document images. The proposed multi-plane segmentation approach first decomposes the document image into distinct object planes to extract and separate homogeneous objects including textual regions of interest, non-text objects such as graphics and pictures, and background textures. This process consists of two stages—localized histogram multilevel thresholding and multi-plane region matching and assembling. Then a text extraction procedure is applied on the resultant planes to detect and extract textual objects with different characteristics in the respective planes. The proposed approach processes document images regionally and adaptively according to their respective local features. Hence detailed characteristics of the extracted textual objects, particularly small characters with thin strokes, as well as gradational illuminations of characters, can be well-preserved. Moreover, this way also allows background objects with uneven, gradational, and sharp variations in contrast, illumination, and texture to be handled easily and well. Experimental results on real-life complex document images demonstrate that the proposed approach is effective in extracting textual objects with various illuminations, sizes, and font styles from various types of complex document images.     

A New Text Location Approach Based Wavelet

With the advancement of content-based retrieval technology, the importance of semantics for text information contained in images attracts many researchers. An algorithm which will automatically locate the textual regions in the input image will facilitate     

Combining global, regional and contextual features for automatic image annotation

This paper presents a novel approach to automatic image annotation which combines global, regional, and contextual features by an extended cross-media relevance model. Unlike typical image annotation methods which use either global or regional features exclusively, as well as neglect the textual context information among the annotated words, the proposed approach incorporates the three kinds of information which are helpful to describe image semantics to annotate images by estimating their joint probability. Specifically, we describe the global features as a distribution vector of visual topics and model the textual context as a multinomial distribution. The global features provide the global distribution of visual topics over an image, while the textual context relaxes the assumption of mutual independence among annotated words which is commonly adopted in most existing methods. Both the global features and textual context are learned by a probability latent semantic analysis approach from the training data. The experiments over 5k Corel images have shown that combining these three kinds of information is beneficial in image annotation.     

Critical theory and textual computing: Comments and suggestions

Conclusion Donald Bruce and I do speak the same theoretical metalanguage, though I suspect that he is considerably more fluent than I in that tongue. Given the variety of responses, including Bruce's favorable reaction, to my attempts to provoke theoretical debate concerning the nature of electronic text as a new object of research, I am considerably more optimistic than his knowing smile and tears of rage. It is my contention that researchers in textual computing have significant advantages in reconceptualizing text precisely because computing technology shatters the evident surface structures of text. If electronic text is a radically different object of research, then theoretical models of the kind discussed in this volume should have a significant impact on disciplines which are currently debating the nature and limits of textuality.I would like to suggest that we, as specialists in textual computing, should make every effort to combine abstract theoretical considerations with clear efforts towards empirical verification. Maintaining that difficult balance between theory, method, and empirical verification is, in my opinion, one of the central contributions that theory of textual computing can make to critical theory in general. The computing environment provides an ideal testing ground for literary theories by encouraging experimentation and verification using real data, an element that is all too often overlooked by many critical theorists.     

Textual image compression at low bit rates based on region-of-interest coding

In this paper, we deal with those applications of textual image compression where high compression ratio and maintaining or improving the visual quality and readability of the compressed images are of main concern. In textual images, most of the information exists in the edge regions; therefore, the compression problem can be studied in the framework of region-of-interest (ROI) coding. In this paper, the Set Partitioning in Hierarchical Trees (SPIHT) coder is used in the framework of ROI coding along with some image enhancement techniques in order to remove the leakage effect which occurs in the wavelet-based low-bit-rate compression. We evaluated the compression performance of the proposed method with respect to some qualitative and quantitative measures. The qualitative measures include the averaged mean opinion scores (MOS) curve along with demonstrating some outputs in different conditions. The quantitative measures include two proposed modified PSNR measures and the conventional one. Comparing the results of the proposed method with those of three conventional approaches, DjVu, JPEG2000, and SPIHT coding, showed that the proposed compression method considerably outperformed the others especially from the qualitative aspect. The proposed method improved the MOS by 20 and 30 %, in average, for high- and low-contrast textual images, respectively. In terms of the modified and conventional PSNR measures, the proposed method outperformed DjVu and JPEG2000 up to 0.4 dB for high-contrast textual images at low bit rates. In addition, compressing the high contrast images using the proposed ROI technique, compared to without using this technique, improved the average textual PSNR measure up to 0.5 dB, at low bit rates.     

DocMIR: An automatic document-based indexing system for meeting retrieval

This paper describes the DocMIR system which captures, analyzes and indexes automatically meetings, conferences, lectures, etc. by taking advantage of the documents projected (e.g. slideshows, budget tables, figures, etc.) during the events. For instance, the system can automatically apply the above-mentioned procedures to a lecture and automatically index the event according to the presented slides and their contents. For indexing, the system requires neither specific software installed on the presenter’s computer nor any conscious intervention of the speaker throughout the presentation. The only material required by the system is the electronic presentation file of the speaker. Even if not provided, the system would temporally segment the presentation and offer a simple storyboard-like browsing interface. The system runs on several capture boxes connected to cameras and microphones that records events, synchronously. Once the recording is over, indexing is automatically performed by analyzing the content of the captured video containing projected documents and detects the scene changes, identifies the documents, computes their duration and extracts their textual content. Each of the captured images is identified from a repository containing all original electronic documents, captured audio–visual data and metadata created during post-production. The identification is based on documents’ signatures, which hierarchically structure features from both layout structure and color distributions of the document images. Video segments are finally enriched with textual content of the identified original documents, which further facilitate the query and retrieval without using OCR. The signature-based indexing method proposed in this article is robust and works with low-resolution images and can be applied to several other applications including real-time document recognition, multimedia IR and augmented reality systems.

Visual Knowledge: Textual Iconography of the Quixote, a Hypertextual Archive

Ever since its initial publication four hundred years ago, thousandsof editions, most often illustrated, have been published ofCervantes' masterpiece, Don Quixote. Imagery has become an integralpart of the reception and interpretation of the text. To date,a comprehensive collection of these images, the textual iconographyof the Quixote, has not been published. We report in this paperon overcoming two key obstacles: limitations on the availabilityof materials and limitations due to the technical and financialcharacteristics of print-based dissemination. Our digital iconographymakes a rich artistic tradition accessible to readers for thefirst time, and reveals a wealth of information about the historical,cultural, and literary contexts into which the Quixote has beenplaced.     

How to teach your edition how to swim

The mutability of electronic editions confronts editors witha new world, in which large parts of current editorial theorymust be re-thought, based as it often is on assumptions basedon the properties of paper editions. Software can adapt moreeasily than paper to the needs and interests of the reader,which means many choices about the selection of informationin an edition and its presentation to the reader no longer needto be fixed for all time, but can be left open for the reader.Software also tends to have a very short lifetime compared topaper; in order to remain usable for more than a few years,electronic editions must find ways of representing the essentialinformation of the edition in software-independent, non-proprietaryways.     

MC-JBIG2: an improved algorithm for Chinese textual image compression

Standard JBIG2 algorithms for textual image compression focus on the features of alphabetic characters such as English, not considering the features of pictograph characters such as Chinese. In this work, an improved algorithm called MC-JBIG2 is developed, which aims at improving compression ratio for Chinese textual images. In the proposed method, first multiple features are extracted from the characters in the images. After that, a cascade of clusters is introduced to accomplish the pattern-matching task for the characters. Finally, to optimize the parameters used in the cascade of clusters, a Monte Carlo strategy is implemented to traverse the feasible space. Experimental results show MC-JBIG2 outperforms existing representative JBIG2 algorithms and systems on Chinese textual images. MC-JBIG2 can also improve compression ratio on Latin textual images, however, the improvement on Latin textual images is not as stable as the improvement on Chinese ones.     

利用不同尺度下复杂性的差异区分文字和照片

将信号分成多个区域,给予“0”,“1”新的定义,计算不同尺度下的复杂性而不增加符号数目．对文字图像和照片的分析显示,随着尺度的减小,照片复杂性增大的幅度大于文字．该结果说明：不同尺度下复杂性的差异可以作为图像分类的新方法或者作为现有分类器的特征．     

The very pulse of the machine: Three trends toward improvement in electronic versions of humanities texts

Since April 1989, the Center for Text and Technology at Georgetown University has gathered information on the structure of projects that produce electronic text in the humanities. This report — based on the April, 1991 version of the Georgetown Catalogue and emphasizing its full-text projects in humanities disciplines other than linguistics —surveys the countries in which projects are found, the languages encoded, the disciplines served, and the auspices represented. Then the report explores three trends toward the improvement of electronic texts: increased scope of the new projects, improved quality of the editions used, and greater sophistication in the text-analysis tools added. Included among the notes is a list of titles and contacts for 42 projects cited in the report.Michael Neuman is Director of Georgetown University's Center for Text and Technology, whose mission is the creation and dissemination of electronic text for the enhancement of teaching and research in the humanities. He has taught English literature, but his recent articles and presentations focus on electronic editions of philosophical works.James A. Wilderotter 11, Project Assistant at the Center for Text and Technology, has provided many of the compilations in this report and gathered much of the data in the current version of the Georgetown Catalogue of Projects in Electronic Text.     

A Multimodal Scheme for Program Segmentation and Representation in Broadcast Video Streams

With the advance of digital video recording and playback systems, the request for efficiently managing recorded TV video programs is evident so that users can readily locate and browse their favorite programs. In this paper, we propose a multimodal scheme to segment and represent TV video streams. The scheme aims to recover the temporal and structural characteristics of TV programs with visual, auditory, and textual information. In terms of visual cues, we develop a novel concept named program-oriented informative images (POIM) to identify the candidate points correlated with the boundaries of individual programs. For audio cues, a multiscale Kullback-Leibler (K-L) distance is proposed to locate audio scene changes (ASC), and accordingly ASC is aligned with video scene changes to represent candidate boundaries of programs. In addition, latent semantic analysis (LSA) is adopted to calculate the textual content similarity (TCS) between shots to model the inter-program similarity and intra-program dissimilarity in terms of speech content. Finally, we fuse the multimodal features of POIM, ASC, and TCS to detect the boundaries of programs including individual commercials (spots). Towards effective program guide and attracting content browsing, we propose a multimodal representation of individual programs by using POIM images, key frames, and textual keywords in a summarization manner. Extensive experiments are carried out over an open benchmarking dataset TRECVID 2005 corpus and promising results have been achieved. Compared with the electronic program guide (EPG), our solution provides a more generic approach to determine the exact boundaries of diverse TV programs even including dramatic spots.     

Wired women writing: Towards a feminist theorization of hypertext

The electronic classroom provides a space for examining the central debates of contemporary feminism, particularly by applying feminist ideas to a theorization of hypertext and creating what I call feminist activist autobiographical hypertexts. In a feminist electronic classroom, we explore the potential for hypertext as a form with which to interrogate dominant ideologies and to produce alternative knowledge. In hypertexts informed by radical feminist theory, we bridge feminist theorizations of the social constructedness of subjectivity, and especially the mass media's role in such construction, with materialist feminist critiques of late capifalism and its oppressive institutions. We use feminist activist art as a model and take advantage of the way hypertext enables us to combine the best of both modern and postmodern strategies of textual production. Such strategies include employing a multiplicity of perspectives, collage (single-screen juxtapositions of text and image), montage (juxtaposition through linking), the juxtaposition of autobiography with social critique and with a critique of dominant media representations, the reappropriation of mass-mediated images and the recontextualization of dominant ideological signs, the articulation of the voices and experiences of previously silenced female subject, and the combination of conceptual, experiential, and emotional knowledge, including knowledge from the unconscious, where much oppressive ideology lies internalized and often repressed.     

Multimedia ontology matching by using visual and textual modalities

Ontologies have been intensively applied for improving multimedia search and retrieval by providing explicit meaning to visual content. Several multimedia ontologies have been recently proposed as knowledge models suitable for narrowing the well known semantic gap and for enabling the semantic interpretation of images. Since these ontologies have been created in different application contexts, establishing links between them, a task known as ontology matching, promises to fully unlock their potential in support of multimedia search and retrieval. This paper proposes and compares empirically two extensional ontology matching techniques applied to an important semantic image retrieval issue: automatically associating common-sense knowledge to multimedia concepts. First, we extend a previously introduced textual concept matching approach to use both textual and visual representation of images. In addition, a novel matching technique based on a multi-modal graph is proposed. We argue that the textual and visual modalities have to be seen as complementary rather than as exclusive sources of extensional information in order to improve the efficiency of the application of an ontology matching approach in the multimedia domain. An experimental evaluation is included in the paper.