Search

Processing and analyzing historical manuscripts is considered one of the most challenging problems in the document analysis and recognition domain. Manuscripts written in cursive are even more difficult due to overlapping words with random spacing, irregular and varying characters' shapes, poor scan quality, and insufficient labeled data. Despite the significant achievements of deep learning approaches in computer vision, handwritten word recognition is far from solved. Most of the existing methods focus on well-segmented word datasets. In this paper, we present an empirical study investigating how well state-of-the-art deep learning models perform on detection and recognition of handwritten words in Spanish American notary records. Professional historians were involved in preparing a labeled dataset of 26,482 Spanish words employed in the experiments. We investigate the performance of some state-of-the-art models on optical character recognition (OCR) on handwritten text documents: Keras-OCR, the object detection algorithm "You Only Look Once" (YOLO), Tesseract OCR, Kraken, and Calamari-OCR. Since YOLO does not include a text recognizer, we propose YOLO-OCR, an innovative model to detect and recognize words in historical manuscripts written in Spanish. Our results show the performance of pre-trained models on our dataset and that Keras-OCR and YOLO-OCR models are highly valuable for content extraction.

Although numerous research studies have been conducted concerning automatic vehicle number plate detection and recognition, various presented automated number plate recognition systems are devised for specific countries where number plates follow standard patterns. However, such systems cannot be applied in Afghanistan because of the different designs and the language. Moreover, due to the cursive nature, writing direction, and shape variation of the Pashto characters, the segmentation of words into isolated characters is a more complicated task. Hence, the Pashto optical character recognition is a less developed area. To date, no research study has been conducted for Afghanistan number plate detection and recognition. The details on the Afghanistan number plate include character, numbers, and each province's name. The paper presents the study of its type attempting to detect the number plate from the vehicle image and then recognize the province's name, characters, and numbers on the number plate. In particular, the new method incorporating four core steps. The first step is number plate detection applying canny edge detection based on user-defined thresholding and extracts the number plate involving several image processing techniques. The second phase is number plate adjustment using Randon transform-based techniques. The third stage is number plate segmentation isolating each character, number, and province name on the number plate using a scanning approach. The final step employs a convolutional neural network to classify the number plate's alphanumeric characters and provinces' names. In addition, two datasets have been created: the dataset for alphanumeric characters contains 2800 images of 14 classes, and the dataset for provinces' names contains 6800 images of 34 classes. The proposed models present 99.93 percent accuracy for provinces' names classification and 98.93 percent for alphanumeric characters' classification.

In this study, we approached Japanese cursive character (kuzushiji) recognition using machine learning to improve the efficiency of their transcription process. In particular, we discussed methods that automatically distinguish difficult-to-recognize characters or characters that were not included in the training, and put them on hold as unknown characters. We analyzed two methods that deferred recognition of characters: In the first method, characters with low confidence values obtained from the recognition model were put on hold as unknown characters. In the second method, we trained a “miscellaneous class” that combined multiple character classes and deferred the recognition of characters if they were recognized as belonging to the “miscellaneous class.” Experiments showed that by narrowing down the characters to be recognized, a recognition rate of more than 90% was achieved for over 1,500 character classes, confirming the usefulness of temporarily deferring recognition during the transcription process.

Cursive writing is an activity writing beginnings in Primary school. Students in Primary school when learning to write cursive often experience problems berupa the writing is still not neat, the legibility is not clear and not fluent in writing capital letters, and the writing is not in accordance with the lines of the book. The lack of student ability is due to a lack of knowledge in writing rules and proper methods of writing cursive. The purpose of this research is know the method learning to write cursive used in the 2013 curriculum at the State Primary School in Sragen in and method proper learning for students of the existing methods. The method used in this research is descriptive method with a qualitative approach in primary schools in Sragen. The research subjects were second grade students of primary School. The sampling technique was carried out by purposive sampling. Data collection techniques through interviews, observation and documentation. The analysis technique uses an interactive analysis model which consists of three stages, namely: data reduction, data presentation, and conclusion drawing. The results showed that in learning activities to write cursive writing, most teachers apply several methods such as demonstrations, groups/ discussions, assignments, picture and picture, complete sentences, SAS, and Drill. The most appropriate and effective Drill method to be applied in learning because students can practice cursive so that they can form a good habit of writing cursive in students.

Calligraphy is an important part of traditional Chinese culture. Authentic calligraphy works are usually preserved on paper, bamboo slips and stone tablets, which are easy to be destroyed and not easily appreciated by most calligraphy lovers anytime and anywhere. Therefore, in order to facilitate the preservation and appreciation of calligraphy works, digital technology and computer technology are used for digital storage, management and service. The time complexity of image recognition algorithm based on traditional method is relatively high, especially the algorithm based on feature matching has low tolerance to Chinese character deformation. Therefore, this paper tries to use deep learning to study the image recognition of Chinese characters. However, some existing data sets of calligraphic images have a small number of corresponding images, which cannot support deep learning network training. In view of this problem, a data augmentation method based on character glyph and stroke characteristics is proposed. This method starts from two aspects of Chinese character glyph and stroke. Chinese character glyph is changed by affine changes of the restricted conditions. The stroke information of Chinese characters is extracted by super pixel and the stroke information is changed to expand the diversity of Chinese characters in the data set. In the process of character recognition, mixup data augmentation algorithm is combined with Inception v4 network which has excellent performance in Imagenet classification tasks?for network recognition training to improve network generalization ability. Experiments show that the algorithm proposed in this paper based on deep learning has a higher recognition effect than previous algorithms, especially in cursive and running recognition. The digital calligraphy knowledge service system based on web terminal and WeChat small program terminal is designed and implemented, which can provide the digital calligraphy knowledge service of calligraphy work segmentation and calligraphy character recognition.

Chinese cursive characters written in old books are more difficult to recognize than other Chinese Characters such as handwritten Chinese character because they have many various styles. For this reason, it needs a software-based recognition model or service that can recognize Chinese cursive character more easily to assist and support many related researchers. In this paper, we will show the performance analysis results of Chinese cursive character recognition based on Convolutional Neural Network (CNN). In the performance evaluation, we used a dataset collected directly from many old books. Data augmentation technique was applied because our dataset is relatively small dataset when compared to other deep learning datasets. Our optimized recognition model showed about 91.6% recognition accuracy.

Inertial sensors have been used for tracking applications on every scale for decades. Commercial IMUs suffer from a spectrum of errors such as axes misalignment, bias etc. leading to a drifted output. Approximation models are usually implemented to address these issues which are highly sensor-dependent with constrained testing and sometimes use complementary infrastructure which limits their use in the wild. In this demonstration we introduce 'AiRite', an effective solution for 3-D tracking of a smart device using only the onboard IMU. Trajectories of basic shapes and cursive words written in air using smart devices are visualized in 3-D, which are first observations of their kind in the field. We demonstrate device-independence and ubiquity of our tracking method by using different smartphones and smart-watches for writing in air.

The efficiency of any machine learning and computer vision system depends largely on the robustness of feature extraction and selection process. In word spotting applications, many appropriate features have been proposed over the years in literature. Most of these features are extracted for Latin text but are used with Oriental script as well. Extracting features that are more specific to Oriental text is also being investigated and a lot of research is being focused on this aspect lately as well. Deep Learning has also been employed for this purpose. In this paper, we have tried investigate the performance of shape based features for Urdu script. Urdu and Arabic belong to the same family of script and both share similar set of alphabet. This means that features investigated on Urdu will give similar performance for Arabic as well as other Oriental scripts. For this paper, we have compiled results on approximately 21000 ligatures belonging to 200 unique classes taken from scanned pages of the popular Urdu series 'Zaawiyya'. This is Higher Education Commission granted project, due to this data set is provided by them. Proposed system gives encouraging results with precision of 88.5% and recall rate of 90.8%.

One of the important aspects in graphic design is choosing the font of the caption that matches aesthetically the associated image. To obtain a good match, users would exhaustively examine a long font list requiring them a substantial effort. This paper presents FontMatcher, which supports users to design digital graphic works harmoniously pairing fonts with an image. The system provides three features, recommendation, explaination and feedback. If a warm feeling image is given as input, the system recommends warm feeling fonts, and then explains what is the distinguishing features of the recommendation, e.g. a cursive shape. Users can also provide feedback to find fonts which correspond to their intention. Our evaluation results show that the recommended fonts scored better than selected fonts by novices and provides competing results with the ones chosen by experienced graphic designers. The system also provides explanations that help increasing the reliability of the recommended results.

Convolutional Neural Networks (CNNs) have produced excellent results in natural scene semantic pixel labeling tasks. We examine the application of this idea to document processing, using fully supervised Deep CNN semantic segmentation to separate content layers from historical document images containing diverse content types, including handwriting, machine print, form lines, and stamps. For efficiency, we employ a downsampling-upsampling network to make dense pixel predictions. CNNs achieve high generalization accuracy on document images with interleaved, overlapping strokes, even when trained on a solitary pixel-labeled document image. We also show a proof-of-concept extension of the semantic segmentation task to handwritten cursive character recognition, enabling a new "segmentation-free" approach to handwriting transcription.

Nowadays, Arabic handwriting recognition is an active research area. The optical character recognition is classified into two approaches offline and online. There are many studies and applications for Arabic offline recognition, both typed and handwritten, yet there are few studies on Arabic Online recognition. Online recognition, in general, is oriented to only handwritten. The cursive, shapes, dots and delayed strokes of Arabic letters are the most challenging tasks to develop and improve an online system for the Arabic language. Moreover, handwriting of many Arab people becomes poor with low handwriting skills, especially after the cancellation of the Arabic calligraphy subject in the educational system in many Arabic countries. This paper presents a comprehensive survey on Arabic online handwriting recognition for the past few years. The paper aims to elevate the research in this subject, reveal the avenues for improving the recognition of Arabic online handwriting and enhance the skills of the Arab people in handwriting via online teaching and training system.

Recognition of unconstrained handwritten texts is always a difficult problem, particularly if the style of handwriting is a mixed cursive one. Among various Indian scripts, only Bangla has this additional difficulty of tackling mixed cur-siveness of its handwriting style in the pipeline of a method towards its automatic recognition. A few other common recognition difficulties of handwriting in an Indian script include the large size of its alphabet and the extremely cursive nature of the shapes of its alphabetic characters. These are among the reasons of achieving only limited success in the study of unconstrained handwritten Bangla text recognition. Artificial Neural Network (ANN) models have often been used for solving difficult real-life pattern recognition problems. Recurrent Neural Network models (RNN) have been studied in the literature for modeling sequence data. In this study, we consider Long Short Term Memory (LSTM) network model, a useful member of this family. In fact, Bidirectional Long Short-Term Memory (BLSTM) neural networks is a special kind of RNN and have recently attracted special attention in solving sequence labelling problems. In this article, we present a BLSTM architecture based approach for unconstrained online handwritten Bangla text recognition.

International audience; Speech text in comic books is placed and written in a particular manner by the letterers which raises unusual challenges for text recognition. We first detail these challenges and present different approaches to solve them. We compare the performances of generic versus specifically trained OCR systems for typewritten and handwritten text lines from French comic books. This work is evaluated over a subset of public (eBDtheque) and private (Sequencity) datasets. We demonstrate that generic OCR systems perform best on typewritten-like and lowercase fonts while specifically trained OCR can be very powerful on skewed, uppercase and even cursive fonts.

As small displays on devices like smartwatches become increasingly common, many people have difficulty reading the text on these displays. Vision conditions like presbyopia that result in blurry near vision make reading small text particularly hard. We design multiple different scripts for displaying English text, legible at small sizes even when blurry, for small screens such as smartphones and smartwatches. These "smartfonts" redesign visual character presentations to improve the reading experience. Like cursive, Grade 1 Braille, and ordinary fonts, they preserve orthography and spelling. They have the potential to enable people to read more text comfortably on small screens, e.g., without reading glasses. To simulate presbyopia, we blur images and evaluate their legibility using paid crowdsourcing. We also evaluate the difficulty of learning to read smartfonts and observe a learnability/legibility trade-off. Our most learnable smartfont can be read at roughly half the speed of Latin after two thousand practice sentences. It is also legible smaller than half the size of traditional Latin (i.e. "English") when blurry.

Urdu handwritten Document segmentation can be defined as the process of splitting the images of handwritten text document into pieces corresponding to lines, words or ligatures and character. This is a very challenging task because in Urdu handwritten documents, skewed text lines appear frequently with multiple angles. After segmentation of ligature or stroke, also defined as connected components in handwritten text document, we have to sequence the ligatures as appearing in the document so that the meaning of the document is preserved. In this paper, We use bottom up grouping approach for the segmentation. We have used a novel connectivity strength parameter with depth first search approach for extraction of connected components of the same line from complete connected components of the given cursive text document. The exact sequence of connected components is stored in the sequential vector which contains the label of the connected components. The proposed urdu ligature segmentation technique is implemented and tested on various in house generated database having different skewed and curved documents. Quantitative analysis also shows that this approach gives better results and overcomes the problems encountered in Hill-and dale writing styles.

In Chinese calligraphy, the concept of promoting the circulation of qi enables Chinese characters to express the essence, qi, and spirit of the calligrapher through the proper use of force and speed. As a result, the emotions and soul of the calligrapher are recorded in a work of Chinese calligraphy. In Cursive, a performance by internationally renowned Taiwanese dance company, Cloud Gate Dance Theater, the dancers try to figure out the copybook works of famous ancient Chinese calligraphers. Through dance, they attempt to reproduce the essence, qi, and spirit of these calligraphers along with the qi and rhythm within Chinese calligraphy, and re-integrate them with the force and beauty of modern dance. Focusing on reproducing the essence, qi, and spirit of a calligrapher, the interactive mechanical installation work, The Qi of Calligraphy, utilizes clear imagery of copybook works by famous ancient Chinese calligraphers. Operated by a robotic arm and a brush-like extension, a spotlight moves and shifts to reproduce the works of a copybook. With the form and rhythm of a trajectory, it depicts the copybook, jing (strength). Playing with the works game-like features, viewers utilize both hands and their bodies to chase after the trajectory of the spotlight of the mechanical arm, resulting in a dance along with the rhythm of the copybook. And, when viewers are able to touch the spotlight, the image of their bodies are imprinted onto the luminous wall, enabling the points in Chinese character to be displayed on the wall. In the end, the relationship between viewers bodies and the machine is presented along with an intertwining of their bodies with the copybook. As a result, body and qi are merged together.

Person identification and verification based on handwriting is a hot research topic. The use of computer in this area is within recent years. It has received its interest due to a great importance in criminal justice based on handwriting analysis. This paper presents a current research in person identification and verification based on natural as well as disguised and forged handwritings. The focus is on the approaches present for off line cursive and isolated English handwriting up till date. A comparative study is presented for text dependent and text independent methods. It briefly discuss about the software tools available and its applicability. Also gives a summary of research results. It outlines the directions for further research.

With the increasing popularity of pen-based digital devices, online handwriting recognition has generated potential markets in India. Hand-held devices equipped with pen-based technologies are now affordable to a large section of Indian population. However, till date little research has been done on online handwriting recognition of an Indian script. On the other hand, India is a multilingual country and Bangla is its second most popular language used by nearly 220 million people in India and Bangladesh. Unconstrained handwriting in Bangla is cursive in nature unlike other Indian scripts. Difficulties of designing a recognition system for unconstrained Bangla handwriting are mainly due to the fact that Bangla has a very large alphabet set consisting of nearly 300 shapes many of which are very complex and also there are a large number of shape similar characters. In this article, we describe preliminary results of our recent study on limited vocabulary Bangla cursive handwriting recognition based on somewhat unusual combination of multilayer perceptron (MLP) and support vector machine (SVM) classifiers. We simulated the proposed approach on two vocabularies of sizes 50 and 110 and the recognition performance on these two are comparable.

International audience; In this paper, we present a new type of Marking menus. Continuous Marking Menus are specifically dedicated to pen-based interfaces, and designed to define a set of cursive, realistic handwritten gestures. In menu mode, they offer a continuous visual feedback and fluent exploration of menu hierarchy, inviting the user to execute cursive gestures for invoking the desired commands. In marking mode, a specific gesture recognition method is proposed and proved to be very efficient for recognizing cursive gestures.

This article describes a recent effort for database creation and recognition of online handwritten isolated basic characters of Bangla, the second most popular script of the Indian subcontinent. It describes a scheme for extraction of sub-strokes from the online samples of handwritten Bangla characters, which are significantly cursive in shapes. The proposed recognition scheme includes a new feature vector to be computed for each sub-stroke. The recognition performance of the test samples of the present database is evaluated separately on two classifiers -- a Hidden Markov Model (HMM) based classifier and a nearest-neighbour classifier based on Dynamic Time Warping (DTW). The second classifier outperforms the HMM-based classifier for the present test set.

In this paper, we describe an approach for the problem of segmenting overlapping characters. We are working with digit segmentation for bank check processing. Our method is based on the idea of a hypothetical ball traversing the number. The inertia of the movement segments the overlapping digits. Rules are defined for this movement. Our initial proposal achieved very good results with O(n2) complexity.

Nastaliq is a calligraphic, beautiful and more aesthetic style of writing Urdu, the national language of Pakistan, also used to read and write in India and other countries of the region.OCRs developed for many world languages are already under efficient use but none exist for Nastaliq -- a calligraphic adaptation of the Arabic scrip which is inherently cursive in nature.In Nastaliq, word and character overlapping makes optical recognition more complex.This paper presents the ongoing research on Nastaliq Optical Character Recognition (NOCR). In this research, we have proposed a novel segmentation-free technique for the design and implementation of a Nastaliq OCR based on cross-correlation.

In this paper we present a novel system for sketching the motion of a character. The process begins by sketching a character to be animated. An animated motion is then created for the character by drawing a continuous sequence of lines, arcs, and loops. These are parsed and mapped to a parameterized set of output motions that further reflect the location and timing of the input sketch. The current system supports a repertoire of 18 different types of motions in 2D and a subset of these in 3D. The system is unique in its use of a cursive motion specification, its ability to allow for fast experimentation, and its ease of use for non-experts.

We describe an interaction technique for expressive nonverbal communication through an avatar. The input consists of letters written with expressive modulation, by hand, using a pen tablet input device. This technique, which we call Cursive, translates an alphabetic letter into a body gesture. Our design is inspired by the observation that handwriting, like some kinds of human gesture, informs through the symbols it depicts, and through the manner in which the symbols are produced. The technique is especially suited for desktop and portable virtual reality applications--applications that do not use special rooms or equipment but only two-dimensional displays and simple pointing devices for interaction with the virtual environment. Our system solves the difficult problem of controlling the movements of a highly articulated, 3D avatar using a common input device within the context of an office or other real world environment.

We are developing an interaction technique for rich nonverbal communication through an avatar. By writing a single letter on a pen tablet device, a user can express their ideas or intentions, non-verbally, using their avatar body. Our system solves the difficult problem of controlling the movements of a highly articulated, 3D avatar model using a common input device within the context of an office environment. We believe that writing is a richly expressive and natural means for controlling expressive avatar gesture.