Academic literature on the topic 'Hand-written text ; Cursive script recognition'

Currently, no offline tool is available for Optical Character Recognition (OCR) in Kurdish. Kurdish is spoken in different dialects and uses several scripts for writing. The Persian/Arabic script is widely used among these dialects. The Persian/Arabic script is written from Right to Left (RTL), it is cursive, and it uses unique diacritics. These features, particularly the last two, affect the segmentation stage in developing a Kurdish OCR. In this article, we introduce an enhanced character segmentation based method which addresses the mentioned characteristics. We applied the method to text-only images and tested the Kurdish OCR using documents of different fonts, font sizes, and image resolutions. The results of the experiments showed that the accuracy rate of character recognition of the proposed method was 90.82% on average.


 
 
 Segmentation is an important step for developing any optical character recognition (OCR) system, which has to be redesigned for each script having, non-uniform nature/property. It is used to decompose the image into its sub-units, which act as a basis for character recognition. Brahmi is a non-cursive ancient script, in which characters are not attached to each other and have some spacing between them. This study analyses various segmentation methods for different scripts to develop the best suitable segmentation method for Brahmi. MATLAB software was used for segmentation purpose in the experiment. The sample data belongs to Brahmi script-based ‘Rumandei inscription’. In this paper, we discuss a segmentation methodology for distinct components, namely text lines, words and characters of Rumandei inscription, written in Brahmi script. For segmenting distinct components of inscription different approach were used like horizontal projection profile, vertical projection profile and Relative minima approach. This is fundamental research on an inscription based on Brahmi script, which acts as a foundation for developing a segmentation module of an OCR solution/system of similar scripts in future. Information search and retrieval is an important activity of a library. So, to ensure this support for digitised documents written in ancient script, their character recognition is mandatory through the OCR system.
 
 

There are two main techniques to convert written or printed text into digital format. The first technique is to create an image of written/printed text, but images are large in size so they require huge memory space to store, as well as text in image form cannot be undergo further processes like edit, search, copy, etc. The second technique is to use an Optical Character Recognition (OCR) system. OCR’s can read documents and convert manual text documents into digital text and this digital text can be processed to extract knowledge. A huge amount of Urdu language’s data is available in handwritten or in printed form that needs to be converted into digital format for knowledge acquisition. Highly cursive, complex structure, bi-directionality, and compound in nature, etc. make the Urdu language too complex to obtain accurate OCR results. In this study, supervised learning-based OCR system is proposed for Nastalique Urdu language. The proposed system evaluations under a variety of experimental settings apprehend 98.4% training results and 97.3% test results, which is the highest recognition rate ever achieved by any Urdu language OCR system. The proposed system is simple to implement especially in software front of OCR system also the proposed technique is useful for printed text as well as handwritten text and it will help in developing more accurate Urdu OCR’s software systems in the future.

THE AUTOGRAPH OF BOGUSLAVAS RADVILA IN THE CONTEXT OF THE 17TH-CENTURY ROMAN CURSIVE The article dwells on the autograph of Boguslavas Radvila (Bogusław Radziwiłł) (1620–1669). The earliest autograph of the duke in the analysed material is detected in two letters in the Polish language dating back to 1622, and one letter in the same language dating back to 1623, written in his name by an adult to Kristupas II Radvila (Krzysztof Radziwiłł). The assumption is that the validation ręką swą in the letters of 1622 and the subscription, signature and validation in the letter of 1623 were made by the young duke while his hand was being held by an adult. The slightly different ductus of Boguslavas’ ‘autograph’, as observed in the letters, suggests that this might have been two different adults. It still remains unclear whether Boguslavas, who grew up in his mother’s German household and only spoke German, was taught to write in the German Gothic Cursive. While in the custody of his uncle Kristupas II Radvila, from 1629 to 1633, Boguslavas Radvila was taught to write, and used four different version of Italian Humanist Bastard. Based on the chronology, these versions are conditionally referred to as the first, the second, the third and the fourth. Letters in Polish to his uncle dating from 1629 and 1630 were written in the first and second versions. Letters to his uncle in Latin dating from 1630–1633 were written or signed (if they were written by somebody else) in the third and fourth versions. For a child’s hand, most letters are written calligraphically. The abundance of decorative elements suggests that Boguslavas was increasingly in control of the quill, and that he had talent. In addition, it points to the fact that when teaching a nobleman to write, the focus was on calligraphy. It still remains unclear why the letters in Latin were written in different versions of Humanist Bastard to those in Polish. The cautious assumption can be made that this was due to the preference of the duke’s teachers for those particular versions of testeggiata cursive. Consistent changes in Radvila’s autograph from a child’s to an adult’s hand could not be traced. Undoubtedly, the autograph was influenced by his studies between 1637 and 1648, and his visits to countries (Germany, the Netherlands, France and England) where the national script was Gothic Cursive, though Italian Humanist Bastard and French Humanist Bastard were also used. By 1641, the duke’s hand had already changed. The autograph was half-cursive, and included Gothic forms of d and z which had not been used in childhood. By 1645, Boguslavas Radvila’s hand had been fully trained, and the autograph had fully formed in the shape of cursive ductus. The choice of cursive was subject to the language of the text. Accuracy when writing was also an important factor. Based on the nature of the cursive used, four versions of the nobleman’s autograph can be distinguished: a Humanistic, a Mixed one, and two Gothic, one German Gothic and the other French Gothic. Texts in Polish and Latin (insertions in the text) were written either in Humanistic or in Mixed Cursive. Italian Humanistic Bastard served as the basis for the Humanistic autograph. However, quite a few of its characteristic features had paled, had been modified, or substituted with elements of French Humanistic Bastard, Gothic or Gothicised characters. The more accurate the duke was, the more elements of Italian Humanistic Bastard could be detected in his autograph. Occasionally, some features of French Humanistic Bastard can be identified (a letter or two). When writing in a fast and casual manner, the number of Gothic and Gothicised elements increased, which made the cursive Mixed. Texts in French were written in Gothic French Cursive. When the duke was not accurate enough, the number of Humanistic elements in his French Gothic autograph increased. This was Mixed Cursive with elements of French Gothic Cursive. Texts in German were written in German Gothic Cursive, under the influence of Humanistic Cursive. This is obvious in the ductus and occasional Humanistic form of the letters. Boguslavas Radvila’s autograph was subject to change: it either became neater or more casual. The calligraphic, or close to calligraphy, and casual, at times almost illegible, version of his cursive can be distinguished. Differences in the autograph can be explained by the duke’s attitude when writing, and by his physical well-being. Boguslavas Radvila’s autograph is evidence of the two types of Roman Cursive that were used at that time. At the same time, the duke joins the ranks of those who mastered two different cursives.

Handwritten character recognition is increasingly important in a variety of automation fields, for example, authentication of bank signatures, identification of ZIP codes on letter addresses, and forensic evidence. Despite improved object recognition technologies, Pashto’s hand-written character recognition (PHCR) remains largely unsolved due to the presence of many enigmatic hand-written characters, enormously cursive Pashto characters, and lack of research attention. We propose a convolutional neural network (CNN) model for recognition of Pashto hand-written characters for the first time in an unrestricted environment. Firstly, a novel Pashto handwritten character data set, “Poha”, for 44 characters is constructed. For preprocessing, deep fusion image processing techniques and noise reduction for text optimization are applied. A CNN model optimized in the number of convolutional layers and their parameters outperformed common deep models in terms of accuracy. Moreover, a set of benchmark popular CNN models applied to Poha is evaluated and compared with the proposed model. The obtained experimental results show that the proposed model is superior to other models with test accuracy of 99.64 percent for PHCR. The results indicate that our model may be a strong candidate for handwritten character recognition and automated PHCR applications.

Handwriting Detection is a process or potential of a computer program to collect and analyze comprehensible input that is written by hand from various types of media such as photographs, newspapers, paper reports etc. Handwritten Text Recognition is a sub-discipline of Pattern Recognition. Pattern Recognition is refers to the classification of datasets or objects into various categories or classes. Handwriting Recognition is the process of transforming a handwritten text in a specific language into its digitally expressible script represented by a set of icons known as letters or characters. Speech synthesis is the artificial production of human speech using Machine Learning based software and audio output based computer hardware. While there are many systems which convert normal language text in to speech, the aim of this paper is to study Optical Character Recognition with speech synthesis technology and to develop a cost effective user friendly image based offline text to speech conversion system using CRNN neural networks model and Hidden Markov Model. The automated interpretation of text that has been written by hand can be very useful in various instances where processing of great amounts of handwritten data is required, such as signature verification, analysis of various types of documents and recognition of amounts written on bank cheques by hand.

Optical Character Recognition has emerged as an attractive research field nowadays. Lot of work has been done in Urdu script based on various approaches and diverse methodologies have been put forward based on Nastaliq font style. Urdu is written diagonally from top to bottom, the style known as Nastaliq. This feature of Nastaliq makes Urdu highly cursive and more sensitive leading to a difficult recognition problem. Due to the peculiarities of Nastaliq Style of writing, we have chosen ligature as a basic unit of recognition in order to reduce the complexity of system. The accuracy rate of recognizing ligature in Urdu text corresponds to the efficiency with which the ligatures are segmented. In addition to extracting connected components, the ligature segmentation takes into consideration various factors like baseline information, height, width, and centroid. In this paper ligature Recognition is performed by using multi-SVM (Sup-port Vector Machine) approach which gives an accuracy of 97% when 903 text images are fed to it.

The need for offline handwritten character recognition is intense, yet difficult as the writing varies from person to person and also depends on various other factors connected to the attitude and mood of the person. However, we are able to achieve it by converting the handwritten document into digital form. It has been advanced with introducing convolutional neural networks and is further productive with pre-trained models which have the capacity of decreasing the training time and increasing accuracy of character recognition. Research in recognition of handwritten characters for Indian languages is less when compared to other languages like English, Latin, Chinese etc., mainly because it is a multilingual country. Recognition of Telugu and Hindi characters are more difficult as the script of these languages is mostly cursive and are with more diacritics. So the research work in this line is to have inclination towards accuracy in their recognition. Some research has already been started and is successful up to eighty percent in offline hand written character recognition of Telugu and Hindi. The proposed work focuses on increasing accuracy in less time in recognition of these selected languages and is able to reach the expectant values.

Handwriting recognition has been one of the active and challenging research areas in the field of pattern recognition. It has numerous applications which include, reading aid for blind, bank cheques and conversion of any hand written document into structural text form[1]. As there are no sufficient number of works on Indian language character recognition especially Kannada script among 15 major scripts in India[2].In this paper an attempt is made to recognize handwritten Kannada characters using Feed Forward neural networks. A handwritten kannada character is resized into 20x30 pixel.The resized character is used for training the neural network. Once the training process is completed the same character is given as input to the neural network with different set of neurons in hidden layer and their recognition accuracy rate for different kannada characters has been calculated and compared. The results show that the proposed system yields good recognition accuracy rates comparable to that of other handwritten character recognition systems.

As the ball dropped on 1999, is it any wonder that No Doubt played, “It’s the End of the World as We Know It” by R.E.M. live on MTV? Any discussion of the Nineties—and its pinnacle moment, Y2K—requires a discussion of both the cover and the glitch, two performative and technological enactments that fomented the collapse between author-reader and user-machine that has, twenty years later, become normalised in today’s Post Internet culture. By staging failure and inviting the audience to participate, the glitch and the cover call into question the original and the origin story. This breakdown of normative borders has prompted the convergence of previously demarcated media, genres, and cultures, a constellation from which to recognise a stochastic hybrid form. The Cover as a Revelation of Collaborative MurmurBefore Sean Parker collaborated with Shawn Fanning to launch Napster on 1 June 1999, networked file distribution existed as cumbersome text-based programs like Internet Relay Chat and Usenet, servers which resembled bulletin boards comprising multiple categories of digitally ripped files. Napster’s simple interface, its advanced search filters, and its focus on music and audio files fostered a peer-to-peer network that became the fastest growing website in history, registering 80 million users in less than two years.In harnessing the transgressive power of the Internet to force a new mode of content sharing, Napster forced traditional providers to rethink what constitutes “content” at a moment which prefigures our current phenomena of “produsage” (Bruns) and the vast popularity of user-generated content. At stake is not just the democratisation of art but troubling the very idea of intellectual property, which is to say, the very concept of ownership.Long before the Internet was re-routed from military servers and then mainstreamed, Michel Foucault understood the efficacy of anonymous interactions on the level of literature, imagining a culture where discourse would circulate without any need for an author. But what he was asking in 1969 is something we can better answer today, because it seems less germane to call into question the need for an author in a culture in which everyone is writing, producing, and reproducing text, and more effective to think about re-evaluating the notion of a single author, or what it means to write by yourself. One would have to testify to the particular medium we have at our disposal, the Internet’s ultimate permissibility, its provocations for collaboration and co-creation. One would have to surrender the idea that authors own anything besides our will to keep producing, and our desire for change; and to modulate means to resist without negating, to alter without omitting, to enable something new to come forward; the unfolding of the text into the anonymity of a murmur.We should remind ourselves that “to author” all the way down to its Latin roots signifies advising, witnessing, and transferring. We should be reminded that to author something means to forget the act of saying “I,” to forget it or to make it recede in the background in service of the other or others, on behalf of a community. The de-centralisation of Web development and programming initiated by Napster inform a poetics of relation, an always-open structure in which, as Édouard Glissant said, “the creator of a text is effaced, or rather, is done away with, to be revealed in the texture of his creation” (25). When a solid melts, it reveals something always underneath, something at the bottom, something inside—something new and something that was always already there. A cover, too, is both a revival and a reworking, an update and an interpretation, a retrospective tribute and a re-version that looks toward the future. In performing the new, the original as singular is called into question, replaced by an increasingly fetishised copy made up of and made by multiples.Authorial Effacement and the Exigency of the ErrorY2K, otherwise known as the Millennium Bug, was a coding problem, an abbreviation made to save memory space which would disrupt computers during the transition from 1999 to 2000, when it was feared that the new year would become literally unrecognisable. After an estimated $300 billion in upgraded hardware and software was spent to make computers Y2K-compliant, something more extraordinary than global network collapse occurred as midnight struck: nothing.But what if the machine admits the possibility of accident? Implicit in the admission of any accident is the disclosure of a new condition—something to be heard, to happen, from the Greek ad-cadere, which means to fall. In this drop into non-repetition, the glitch actualises an idea about authorship that necessitates multi-user collaboration; the curtain falls only to reveal the hidden face of technology, which becomes, ultimately, instructions for its re-programming. And even as it deviates, the new form is liable to become mainstreamed into a new fashion. “Glitch’s inherently critical moment(um)” (Menkman 8) indicates this potential for technological self-insurgence, while suggesting the broader cultural collapse of generic markers and hierarchies, and its ensuing flow into authorial fluidity.This feeling of shock, this move “towards the ruins of destructed meaning” (Menkman 29) inherent in any encounter with the glitch, forecasted not the immediate horror of Y2K, but the delayed disasters of 9/11, Hurricane Katrina, Deepwater Horizon Oil Spill, Indian Ocean tsunami, Sichuan Province earthquake, global financial crisis, and two international wars that would all follow within the next nine years. If, as Menkman asserts, the glitch, in representing a loss of self-control “captures the machine revealing itself” (30), what also surfaces is the tipping point that edges us toward a new becoming—not only the inevitability of surrender between machine and user, but their reversibility. Just as crowds stood, transfixed before midnight of the new millennium in anticipation of the error, or its exigency, it’s always the glitch I wait for; it’s always the glitch I aim to re-create, as if on command. The accidental revelation, or the machine breaking through to show us its insides. Like the P2P network that Napster introduced to culture, every glitch produces feedback, a category of noise (Shannon) influencing the machine’s future behaviour whereby potential users might return the transmission.Re-Orienting the Bizarre in Fantasy and FictionIt is in the fantasy of dreams, and their residual leakage into everyday life, evidenced so often in David Lynch’s Twin Peaks, where we can locate a similar authorial agency. The cult Nineties psycho-noir, and its discontinuous return twenty-six years later, provoke us into reconsidering the science of sleep as the art of fiction, assembling an alternative, interactive discourse from found material.The turning in and turning into in dreams is often described as an encounter with the “bizarre,” a word which indicates our lack of understanding about the peculiar processes that normally happen inside our heads. Dreams are inherently and primarily bizarre, Allan J. Hobson argues, because during REM sleep, our noradrenergic and serotonergic systems do not modulate the activated brain, as they do in waking. “The cerebral cortex and hippocampus cannot function in their usual oriented and linear logical way,” Hobson writes, “but instead create odd and remote associations” (71). But is it, in fact, that our dreams are “bizarre” or is it that the model itself is faulty—a precept premised on the normative, its dependency upon generalisation and reducibility—what is bizarre if not the ordinary modulations that occur in everyday life?Recall Foucault’s interest not in what a dream means but what a dream does. How it rematerialises in the waking world and its basis in and effect on imagination. Recall recollection itself, or Erin J. Wamsley’s “Dreaming and Offline Memory Consolidation.” “A ‘function’ for dreaming,” Wamsley writes, “hinges on the difficult question of whether conscious experience in general serves any function” (433). And to think about the dream as a specific mode of experience related to a specific theory of knowledge is to think about a specific form of revelation. It is this revelation, this becoming or coming-to-be, that makes the connection to crowd-sourced content production explicit—dreams serve as an audition or dress rehearsal in which new learning experiences with others are incorporated into the unconscious so that they might be used for production in the waking world. Bert O. States elaborates, linking the function of the dream with the function of the fiction writer “who makes models of the world that carry the imprint and structure of our various concerns. And it does this by using real people, or ‘scraps’ of other people, as the instruments of hypothetical facts” (28). Four out of ten characters in a dream are strangers, according to Calvin Hall, who is himself a stranger, someone I’ve never met in waking life or in a dream. But now that I’ve read him, now that I’ve written him into this work, he seems closer to me. Twin Peak’s serial lesson for viewers is this—even the people who seem strangers to us can interact with and intervene in our processes of production.These are the moments that a beginning takes place. And even if nothing directly follows, this transfer constitutes the hypothesised moment of production, an always-already perhaps, the what-if stimulus of charged possibility; the soil plot, or plot line, for freedom. Twin Peaks is a town in which the bizarre penetrates the everyday so often that eventually, the bizarre is no longer bizarre, but just another encounter with the ordinary. Dream sequences are common, but even more common—and more significant—are the moments in which what might otherwise be a dream vision ruptures into real life; these moments propel the narrative.Exhibit A: A man who hasn’t gone outside in a while begins to crumble, falling to the earth when forced to chase after a young girl, who’s just stolen the secret journal of another young girl, which he, in turn, had stolen.B: A horse appears in the middle of the living room after a routine vacuum cleaning and a subtle barely-there transition, a fade-out into a fade-in, what people call a dissolve. No one notices, or thinks to point out its presence. Or maybe they’re distracted. Or maybe they’ve already forgotten. Dissolve.(I keep hitting “Save As.” As if renaming something can also transform it.)C: All the guests at the Great Northern Hotel begin to dance the tango on cue—a musical, without any music.D: After an accident, a middle-aged woman with an eye patch—she was wearing the eye patch before the accident—believes she’s seventeen again. She enrolls in Twin Peaks High School and joins the cheerleading team.E: A woman pretending to be a Japanese businessman ambles into the town bar to meet her estranged husband, who fails to recognise his cross-dressing, race-swapping wife.F: A girl with blond hair is murdered, only to come back as another girl, with the same face and a different name. And brown hair. They’re cousins.G: After taking over her dead best friend’s Meals on Wheels route, Donna Hayward walks in to meet a boy wearing a tuxedo, sitting on the couch with his fingers clasped: a magician-in-training. “Sometimes things can happen just like this,” he says with a snap while the camera cuts to his grandmother, bed-ridden, and the appearance of a plate of creamed corn that vanishes as soon as she announces its name.H: A woman named Margaret talks to and through a log. The log, cradled in her arms wherever she goes, becomes a key witness.I: After a seven-minute diegetic dream sequence, which includes a one-armed man, a dwarf, a waltz, a dead girl, a dialogue played backward, and a significantly aged representation of the dreamer, Agent Cooper wakes up and drastically shifts his investigation of a mysterious small-town murder. The dream gives him agency; it turns him from a detective staring at a dead-end to one with a map of clues. The next day, it makes him a storyteller; all the others, sitting tableside in the middle of the woods become a captive audience. They become readers. They read into his dream to create their own scenarios. Exhibit I. The cycle of imagination spins on.Images re-direct and obfuscate meaning, a process of over-determination which Foucault says results in “a multiplication of meanings which override and contradict each other” (DAE 34). In the absence of image, the process of imagination prevails. In the absence of story, real drama in our conscious life, we form complex narratives in our sleep—our imaginative unconscious. Sometimes they leak out, become stories in our waking life, if we think to compose them.“A bargain has been struck,” says Harold, an under-5 bit player, later, in an episode called “Laura’s Secret Diary.” So that she might have the chance to read Laura Palmer’s diary, Donna Hayward agrees to talk about her own life, giving Harold the opportunity to write it down in his notebook: his “living novel” the new chapter which reads, after uncapping his pen and smiling, “Donna Hayward.”He flips to the front page and sets a book weight to keep the page in place. He looks over at Donna sheepishly. “Begin.”Donna begins talking about where she was born, the particulars of her father—the lone town doctor—before she interrupts the script and asks her interviewer about his origin story. Not used to people asking him the questions, Harold’s mouth drops and he stops writing. He puts his free hand to his chest and clears his throat. (The ambient, wind-chime soundtrack intensifies.) “I grew up in Boston,” he finally volunteers. “Well, actually, I grew up in books.”He turns his head from Donna to the notebook, writing feverishly, as if he’s begun to write his own responses as the camera cuts back to his subject, Donna, crossing her legs with both hands cupped at her exposed knee, leaning in to tell him: “There’s things you can’t get in books.”“There’s things you can’t get anywhere,” he returns, pen still in his hand. “When we dream, they can be found in other people.”What is a call to composition if not a call for a response? It is always the audience which makes a work of art, re-framed in our own image, the same way we re-orient ourselves in a dream to negotiate its “inconsistencies.” Bizarreness is merely a consequence of linguistic limitations, the overwhelming sensory dream experience which can only be re-framed via a visual representation. And so the relationship between the experience of reading and dreaming is made explicit when we consider the associations internalised in the reader/audience when ingesting a passage of words on a page or on the stage, objects that become mental images and concept pictures, a lens of perception that we may liken to another art form: the film, with its jump-cuts and dissolves, so much like the defamiliarising and dislocating experience of dreaming, especially for the dreamer who wakes. What else to do in that moment but write about it?Evidence of the bizarre in dreams is only the evidence of the capacity of our human consciousness at work in the unconscious; the moment in which imagination and memory come together to create another reality, a spectrum of reality that doesn’t posit a binary between waking and sleeping, a spectrum of reality that revels in the moments where the two coalesce, merge, cross-pollinate—and what action glides forward in its wake? Sustained un-hesitation and the wish to stay inside one’s self. To be conscious of the world outside the dream means the end of one. To see one’s face in the act of dreaming would require the same act of obliteration. Recognition of the other, and of the self, prevents the process from being fulfilled. Creative production and dreaming, like voyeurism, depend on this same lack of recognition, or the recognition of yourself as other. What else is a dream if not a moment of becoming, of substituting or sublimating yourself for someone else?We are asked to relate a recent dream or we volunteer an account, to a friend or lover. We use the word “seem” in nearly every description, when we add it up or how we fail to. Everything seems to be a certain way. It’s not a place but a feeling. James, another character on Twin Peaks, says the same thing, after someone asks him, “Where do you want to go?” but before he hops on his motorcycle and rides off into the unknowable future outside the frame. Everything seems like something else, based on our own associations, our own knowledge of people and things. Offline memory consolidation. Seeming and semblance. An uncertainty of appearing—both happening and seeing. How we mediate—and re-materialise—the dream through text is our attempt to re-capture imagination, to leave off the image and better become it. If, as Foucault says, the dream is always a dream of death, its purpose is a call to creation.Outside of dreams, something bizarre occurs. We call it novelty or news. We might even bestow it with fame. A man gets on the wrong plane and ends up halfway across the world. A movie is made into the moment of his misfortune. Years later, in real life and in movie time, an Iranian refugee can’t even get on the plane; he is turned away by UK immigration officials at Charles de Gaulle, so he spends the next sixteen years living in the airport lounge; when he departs in real life, the movie (The Terminal, 2004) arrives in theaters. Did it take sixteen years to film the terminal exile? How bizarre, how bizarre. OMC’s eponymous refrain of the 1996 one-hit wonder, which is another way of saying, an anomaly.When all things are counted and countable in today’s algorithmic-rich culture, deviance becomes less of a statistical glitch and more of a testament to human peculiarity; the repressed idiosyncrasies of man before machine but especially the fallible tendencies of mankind within machines—the non-repetition of chance that the Nineties emblematised in the form of its final act. The point is to imagine what comes next; to remember waiting together for the end of the world. There is no need to even open your eyes to see it. It is just a feeling. ReferencesBruns, Axel. “Towards Produsage: Futures for User-Led Content Production.” Cultural Attitudes towards Technology and Communication 2006: Proceedings of the Fifth International Conference, eds. Fay Sudweeks, Herbert Hrachovec, and Charles Ess. Murdoch: School of Information Technology, 2006. 275-84. <https://eprints.qut.edu.au/4863/1/4863_1.pdf>.Foucault, Michel. “Dream, Imagination and Existence.” Dream and Existence. Ed. Keith Hoeller. Pittsburgh: Review of Existential Psychology & Psychiatry, 1986. 31-78.———. “What Is an Author?” The Foucault Reader: An Introduction to Foucault’s Thought. Ed. Paul Rainbow. New York: Penguin, 1991.Glissant, Édouard. Poetics of Relation. Trans. Betsy Wing. Ann Arbor: U of Michigan P, 1997.Hall, Calvin S. The Meaning of Dreams. New York: McGraw Hill, 1966.Hobson, J. Allan. The Dream Drugstore: Chemically Altered State of Conscious­ness. Cambridge: MIT Press, 2001.Menkman, Rosa. The Glitch Moment(um). Amsterdam: Network Notebooks, 2011.Shannon, Claude Elwood. “A Mathematical Theory of Communication.” The Bell System Technical Journal 27 (1948): 379-423.States, Bert O. “Bizarreness in Dreams and Other Fictions.” The Dream and the Text: Essays on Literature and Language. Ed. Carol Schreier Rupprecht. Albany: SUNY P, 1993.Twin Peaks. Dir. David Lynch. ABC and Showtime. 1990-3 & 2017. Wamsley, Erin. “Dreaming and Offline Memory Consolidation.” Current Neurology and Neuroscience Reports 14.3 (2014): 433. “Y2K Bug.” Encyclopedia Britannica. 18 July 2018. <https://www.britannica.com/technology/Y2K-bug>.

The automatic recognition of hand-written text has been a goal for over thirty five years. The highly ambiguous nature of cursive writing (with high variability between not only different writers, but even between different samples from the same writer), means that systems based only on visual information are prone to errors. It is suggested that the application of linguistic knowledge to the recognition task may improve recognition accuracy. If a low-level (pattern recognition based) recogniser produces a candidate lattice (i.e. a directed graph giving a number of alternatives at each word position in a sentence), then linguistic knowledge can be used to find the 'best' path through the lattice. There are many forms of linguistic knowledge that may be used to this end. This thesis looks specifically at the use of collocation as a source of linguistic knowledge. Collocation describes the statistical tendency of certain words to co-occur in a language, within a defined range. It is suggested that this tendency may be exploited to aid automatic text recognition. The construction and use of a post-processing system incorporating collocational knowledge is described, as are a number of experiments designed to test the effectiveness of collocation as an aid to text recognition. The results of these experiments suggest that collocational statistics may be a useful form of knowledge for this application and that further research may produce a system of real practical use.