Khurram Khurshid

Word spotting in historical printed documents using shape and sequence comparisons

Information spotting in scanned historical document images is a very challenging task. The joint use of the mechanical press and of human controlled inking introduced great variability in ink level within a book or even within a page. Consequently characters are often broken or merged together and thus become difficult to segment and recognize. The limitations of commercial OCR engines for information retrieval in historical document images have inspired alternative means of identification of given words in such documents. We present a word spotting method for scanned documents in order to find the word images that are similar to a query word, without assuming a correct segmentation of the words into characters. The connected components are first processed to transform a word pattern into a sequence of sub-patterns. Each sub-pattern is represented by a sequence of feature vectors. A modified Edit distance is proposed to perform a segmentation-driven string matching and to compute the Segmentation Driven Edit (SDE) distance between the words to be compared. The set of SDE operations is defined to obtain the word segmentations that are the most appropriate to evaluate their similarity. These operations are efficient to cope with broken and touching characters in words. The distortion of character shapes is handled by coupling the string matching process with local shape comparisons that are achieved by Dynamic Time Warping (DTW). The costs of the SDE operations are provided by the DTW distances. A sub-optimal version of the SDE string matching is also proposed to reduce the computation time, nevertheless it did not lead to a great decrease in performance. It is possible to enter a query by example or a textual query entered with the keyboard. Textual queries can be used to directly spot the word without the need to synthesize its image, as far as character prototype images are available. Results are presented for different documents and compared with other methods, showing the efficiency of our method.

Towards Searchable Digital Urdu Libraries - A Word Spotting Based Retrieval Approach

Libraries in South Asia hold huge collections of valuable printed documents in Urdu and it is of interest to digitize these collections to make them more accessible. The unavailability of an OCR for Urdu however limits the concept of a digital Urdu library to scanning of documents only, offering very limited search facility based on manually assigned tags. We address this issue by proposing a word spotting based keyword search method for information retrieval in digitized collections of printed Urdu documents. The proposed method is based on segmentation of Urdu text in to partial words and representing each partial word by a set of features. To search a specific word (or phrase), the user provides a query in the form of an image. Comparing the features of the partial words in the query image with the ones already indexed, the user is provided with a list of documents containing occurrences of the queried word. The system evaluated on 50 Urdu documents exhibited a recall of 95.17% and a precision of 94.3%.

Fusion of Word Spotting and Spatial Information for Figure Caption Retrieval in Historical Document Images

We present a method for figure caption detection by employing a fusion of several information sources. The evaluation is performed on documents gathered from the collection of the historical medical digital library Medic@. A method based on perceptual grouping simultaneously segments the vertical and horizontal text lines in a page. Spatial relationships between the text lines and the graphics are considered to select a set of caption line candidates. A feature-based wordspotting method is proposed to retrieve the occurrences of word images similar to a given query.Word-spotting is applied to detect the label of the captions, a word like ‘Fig’, ‘FIG’, ‘Figure’ ...followed by the figure number. Combining spatial information and word recognition greatly improve the detection of caption lines. Our initial experiments process more than 300 pages from three different books.

Hand gesture based user interface for computer using a camera and projector

In this paper, we propose a hand gesture based human computer interaction system comprising of a webcam and a pocket projector. The projector projects the display on the wall on any other plain surface. User can interact with the projected screen using his fingertips which are tracked in air by the camera using ‘Camshift’ tracker. A comparative study of different methods of hands and fingertips detection has been made. A robust method has been developed to detect and recognize single stroke gestures traced with fingertips, which are then translated into actions.

Word Spotting Based Retrieval of Urdu Handwritten Documents

Urdu being one of the most popular languages adopted during different swatches of history has a valuable collection of handwritten scripts in different state libraries of South Asia. Digitizing these collections can serve not only to preserve them but also to make them available to general public. Non existence of an Urdu OCR, however, limits the concept of a digital Urdu library to scanning and manual search of documents only. We present a word spotting based search method for Urdu handwritten text. The text is first segmented into partial words and a set of features is computed from each partial word. The user queries the system using word image. The partial words in the query image are then matched with those in the database and the matched partial words are merged into complete words. The proposed method evaluated on 90 handwritten documents reported encouraging precision and recall rates.

A Novel Approach for Word Spotting Using Merge-Split Edit Distance

Edit distance matching has been used in literature for word spotting with characters taken as primitives. The recognition rate however, is limited by the segmentation inconsistencies of characters (broken or merged) caused by noisy images or distorted characters. In this paper, we have proposed a Merge-split edit distance which overcomes these segmentation problems by incorporating a multi-purpose merge cost function. The system is based on the extraction of words and characters in the text and then attributing each character with a set of features. Characters are matched by comparing their extracted feature sets using Dynamic Time Warping (DTW) while the words are matched by comparing the strings of characters using the proposed Merge-Split Edit distance algorithm. Evaluation of the method on 19th century historical document images exhibits extremely promising results.

A survey of camera modules for CubeSats - Design of imaging payload of ICUBE-1

ICUBE-1, the first pico-satellite of the space program ICUBE (Institute of Space Technology Pakistan cubeSat program), is set to be launched in the second quarter of 2013 as a part of the international cubeSat program [1]. One of the objects of ICUBE-1 is to get the students familiarized with the satellite imaging system involving image capture, image analysis, compression, storage and retrieval. The objective is not to get high resolution earth images, but to take low resolution images, store them and successfully send them to ground station. A low resolution camera with low power consumption can be mounted on the satellite to achieve the desired objective. Due to mass and power constraints, communication bandwidth limitations and the passive nature of ICUBE-1 attitude control, the choice of cameras rests very limited. This communication presents a detailed survey of the imaging payloads in general and camera modules in particular, providing guidelines for future miniature satellite developers in choosing a camera module and designing their imaging payloads. The paper will describe the purpose of an imaging payload in miniature satellites, different imaging technologies like CCD & CMOS, camera unit and its interfaces. It will also compare the resolution, power consumption and field of views of different cameras used in miniature satellites and in the end, will also propose a camera module best suited for ICUBE-1.

Institute of Space Technology CubeSat: ICUBE-1 subsystem analysis and design

For a long while, launching satellites for the purpose of research and technology demonstration largely remained with national space agencies and government organizations as the huge funding requirements inhibited the initiation of such projects at university level. It was this idea of providing, at university level, cheap access to space that prompted the design of miniaturized versions of satellites for research purposes. Specifications of CubeSat, a pico-satellite, were defined to provide easy access to space for educational and research institutions. The improvement in engineering technologies and miniaturization of physical components has enabled design, development and launch of such small low-cost spacecrafts and to date, more than 60 universities, institutions and research organizations have taken part in CubeSat program since its inception in 1999[1]. Institute of Space Technology (IST) adopted the concept of CubeSat development by initiating the satellite program, ICUBE. ICUBE is the premier student satellite program of any educational institution/university in Pakistan. The first satellite of this program is named ICUBE-1. Successful launch of ICUBE-1 and establishing its communication link with the ground are the primary goals of this mission. The satellite has a passive attitude control system and will carry a CMOS camera for experimental purposes. In this paper, we will discuss in detail the design philosophy of ICUBE-1, followed by the preliminary design and analysis of all its subsystems. The required testing and technical support facilities are discussed before the final conclusions.12

Keyword Based Information Retrieval System for Urdu Document Images

Various dynasties ruled the Indian sub-continent and left behind enormous and rich cultural heritage that also included intellectually enriched research in the shape of various documents scripted in Urdu. In order to provide efficient access to this knowledge, analysis though digitizing the existing work is the need of hour. In addition to digitization, efficient search mechanisms also need to be implemented to provide users a rapid access to the queried information. In most cases, the digitized documents are complemented by manually assigned tags which not only is a time consuming task but also provides a very limited search facility. Automating the transcription of these documents using Optical Character Recognition (OCR) systems is also challenging due to the very complex cursive nature of Urdu text. To overcome these limitations, a keyword spotting based information retrieval system for document images is introduced in this study. The proposed technique relies on two major modules, document indexing and retrieval. Images of documents are segmented into partial words (ligatures) and identical partial words (PWs) are grouped into clusters. We introduce the concept of considering each (partial) word as a unique shape and a set of shape descriptors is extracted to characterize the PWs. The clusters of PWs are used to index a given set of documents. During retrieval, the query word presented to the system is matched with the clusters in the database and all documents containing instances of the query word are retrieved and presented to the user. The system evaluated on a set of printed Urdu documents in Nastaliq font realized promising precision and recall rates.

Enhancement of Latent Fingerprint Images with Segmentation Perspective

Latent fingerprints are of decisive influence when it comes to identifying suspects, as these are usually encountered at crime scenes. Also, these serve as a compelling evidence in a court of law. Some of the formidable challenges with latent fingerprints are deficient ridge information and poor ridge clarity due to background noise and non-linear distortions. An effective fingerprint enhancement scheme is suggested to meliorate the clarity and continuity of ridges and valleys in a latent fingerprint image. The proposed method deals with both local (minutia or singular points, ridge termination, bifurcation, broken ridges, short ridges, core and delta) and global features (ridge orientation and ridge frequency), while retaining the true ridge-valley structures and removing noise at the same time. The proposed design establishes scalability, accessibility and flexibility.