Suryaprakash Kompalli

Evaluating the use of GPUs in liver image segmentation and HMMER database searches

In this paper we present the results of parallelizing two life sciences applications, Markov random fields-based (MRF) liver segmentation and HMMERs Viterbi algorithm, using GPUs. We relate our experiences in porting both applications to the GPU as well as the techniques and optimizations that are most beneficial. The unique characteristics of both algorithms are demonstrated by implementations on an NVIDIA 8800 GTX Ultra using the CUDA programming environment. We test multiple enhancements in our GPU kernels in order to demonstrate the effectiveness of each strategy. Our optimized MRF kernel achieves over 130× speedup, and our hmmsearch implementation achieves up to 38× speedup. We show that the differences in speedup between MRF and hmmsearch is due primarily to the frequency at which the hmmsearch must read from the GPUs DRAM.

Creation of data resources and design of an evaluation test bed for Devanagari script recognition

The Indian subcontinent has a large number of languages, dialects, and scripts with the Devanagari script being the primary and most widely used of all the scripts. To date, much of the Devanagari optical character recognition (OCR) research has been restricted to a handful of groups. So, techniques have not yet been widely disseminated or evaluated independently and automated evaluation tools are currently not available for lack of a standard representation of ground-truth and result data. A key reason for the absence of sustained research efforts in off-line Devanagari OCR appears to be the paucity of data resources. Ground truthed data for words and characters, on-line dictionaries, corpora of text documents and reliable, standardized statistical analyses and evaluation tools are currently lacking. So, the creation of such data resources will undoubtedly provide a much needed fillip to researchers working on Devanagari OCR. This paper describes a National Science Foundation sponsored project under the International Digital Libraries program to create data resources that will facilitate development of Devanagari OCR technology and provide a standardized test bed and evaluation tools for Devanagari script recognition.

Segmentation of the Liver from Abdominal CT Using Markov Random Field Model and GVF Snakes

Liver segmentation from scans of the abdominal area is an important step in several diagnostic processes. CT scans of the abdominal area contain several organs in close proximity exhibiting similar image characteristics. In this paper, we present preliminary results on an algorithm that uses Markov random fields to obtain an initial contour of the liver. Gradient vector fields (GVF) and active contours are used to refine the initial estimate and segment the liver. Tests are reported on 13 clinical cases using a similarity metric that combines area and space.

Challenges in OCR of Devanagari documents

OCR of Devanagari script presents a wide range of challenges that are not seen in Latin based scripts. This paper outlines the implementation of a neural network based Devanagari OCR. Experimental results on a standard data set are reported and analyzed.

Devanagari OCR using a recognition driven segmentation framework and stochastic language models

This paper describes a novel recognition driven segmentation methodology for Devanagari Optical Character Recognition. Prior approaches have used sequential rules to segment characters followed by template matching for classification. Our method uses a graph representation to segment characters. This method allows us to segment horizontally or vertically overlapping characters as well as those connected along non-linear boundaries into finer primitive components. The components are then processed by a classifier and the classifier score is used to determine if the components need to be further segmented. Multiple hypotheses are obtained for each composite character by considering all possible combinations of the classifier results for the primitive components. Word recognition is performed by designing a stochastic finite state automaton (SFSA) that takes into account both classifier scores as well as character frequencies. A novel feature of our approach is that we use sub-character primitive components in the classification stage in order to reduce the number of classes whereas we use an n-gram language model based on the linguistic character units for word recognition.

Design and comparison of segmentation driven and recognition driven Devanagari OCR

We outline two different techniques for OCR of machine printed, multi-font Devanagari text. In the first design, words are segmented along linear boundaries. Subsequently, classification is performed with the assumption of accurate segmentation. The second approach uses classifiers to obtain preliminary hypothesis for each segment of the word. These results are used to guide further segmentation of certain pieces. While the former technique is segmentation driven, the latter method follows the paradigm of recognition driven segmentation. The two approaches are compared by using a standard data set

Context sensitive labeling of spinal structure in MR images

We present a new method for automatic detection of the lumbar vertebrae and disk structure from MR images. In clinical settings, radiologists utilize several images of the lumbar structure for diagnosis of lumbar disorders. These images are co-registered by technicians and represent orthogonal features of the lumbar region. We combine information from T1W sagittal, T2W sagittal and T2W axial MR images to automatically label disks and vertebral columns. The method couples geometric and tissue property information available from the three types of images with image analysis approaches to achieve 98.8% accuracy for the disk labeling task on a test set of 67 images containing 335 disks.

Tools for enabling digital access to multi-lingual Indic documents

We present methodologies for three important tasks that will eventually enable digital access of multilingual Indian document images. First, we describe several document image analysis techniques necessary to prepare Devanagari document images for OCR. The second task is OCR for machine printed Devanagari words without the help of a lexicon. We describe the OCR methodology and show how it is being extended to other Indian languages. Finally, we describe a versatile platform that facilitates automatic segmentation of document images in multiple Indian languages and an interface to capture the ground truth corresponding to the text. We use transliterated English text and virtual keyboards in a range of Indian languages for this purpose. The multilingual data entry capabilities of the tool and its underlying UNICODE data representation within a structured XML document also allow users to annotate passages of text in one language in other languages using a markup scheme to switch between scripts. Text and annotations are rendered in the appropriate scripts as the text is being annotated, thus providing users prompt and natural feedback. The XML back-end allows meta-data to be recorded describing the annotated document.

Design of a benchmark dataset, similarity metrics, and tools for liver segmentation

Reliable segmentation of the liver has been acknowledged as a significant step in several computational and diagnostic processes. While several methods have been designed for liver segmentation, comparative analysis of reported methods is limited by the unavailability of annotated datasets of the abdominal area. Currently available generic data-sets constitute a small sample set, and most academic work utilizes closed datasets. We have collected a dataset containing abdominal CT scans of 50 patients, with coordinates for the liver boundary. The dataset will be publicly distributed free of cost with software to provide similarity metrics, and a liver segmentation technique that uses Markov Random Fields and Active Contours. In this paper we discuss our data collection methodology, implementation of similarity metrics, and the liver segmentation algorithm.

An OCR Based Approach for Word Spotting in Devanagari Documents

This paper describes an OCR-based technique for word spotting in Devanagari printed documents. The system accepts a Devanagari word as input and returns a sequence of word images that are ranked according to their similarity with the input query. The methodology involves line and word separation, pre-processing document words, word recognition using OCR and similarity matching. We demonstrate a Block Adjacency Graph (BAG) based document cleanup in the pre-processing phase. During word recognition, multiple recognition hypotheses are generated for each document word using a font-independent Devanagari OCR. The similarity matching phase uses a cost based model to match the word input by a user and the OCR results. Experiments are conducted on document images from the publicly available ILT and Million Book Project dataset. The technique achieves an average precision of 80% for 10 queries and 67% for 20 queries for a set of 64 documents containing 5780 word images. The paper also presents a comparison of our method with template-based word spotting techniques.