Abstract

A 2D function, representing a digital image, is a surface in 3D space. Curvature of such a surface can be exploited to detect ridge and valley like features from images. In this paper, we present an analysis of such curvature based ridge and valley detection techniques and come up with a description for the different classes of ridge and valley profiles which can be detected by them. Such an analysis helps in understanding the scope and limitations of the curvature based techniques. As curvature is a measure of ‘bend’in the cross-section profile along a particular direction of the image intensities,the analysisis presented using 1D functions which represent cross-section profiles of ridges and valleys. The classes of profiles which can be detected by a curvature based technique are described in terms of the properties of the second-derivative of the 1D profile function.

Abstract

Automatic identification of a script in a given document image facilitates many important applications such as automatic archiving of multilingual documents, searching online archives of document images and for the selection of script specific OCR in a multilingual environment. In this paper, we present a scheme to identify different Indian scripts from a document image. This scheme employs hierarchical classification which uses features consistent with human perception. Such features are extracted from the responses of a multi-channel log-Gabor filter bank, designed at an optimal scale and multiple orientations. In the first stage, the classifier groups the scripts into five major classes using global features. At the next stage, a sub-classification is performed based on script-specific features. All features are extracted globally from a given text block which does not require any complex and reliable segmentation of the document image into lines and characters. Thus the proposed scheme is efficient and can be used for many practical applications which require processing large volumes of data. The scheme has been tested on 10 Indian scripts and found to be robust to skew generated in the process of scanning and relatively insensitive to change in font size. This proposed system achieves an overall classification accuracy of 97.11% on a large testing data set. These results serve to establish the utility of global approach to classification of scripts.

Abstract

The notion of topographic features like ridges, trenches, hills, etc. is formed by visualising the 2D image function as a surface in 3D space. Hence, properties of such a surface can be used to detect features from images. One such property, the curvature of the image surface, can be used to detect features characterised by a sharp bend in the surface. Curvature based feature detection requires an efficient technique to estimate/calculate the surface curvature. In this paper, we present an alternative measure for curvature and provide an analysis of the same to determine its scope. Feature detection algorithms using this measure are formulated and two applications are chosen to demonstrate their performance. The results show good potential of the proposed measure in terms of efficiency and scope.

Abstract

Boundary detection in natural images is a fundamental problem in many computer vision tasks. In this paper, we argue that early stages in primary visual cortex provide ample information to address the boundary detection problem. In other words, global visual primitives such as object and region boundaries can be extracted using local features captured by the receptive fields. The anatomy of visual cortex and psychological evidences are studied to identify some of the important underlying computational principles for the boundary detection task. A scheme for boundary detection based on these principles is developed and presented. Results of testing the scheme on a benchmark set of natural images, with associated human marked boundaries, show the performance to be quantitatively competitive with existing computer vision approaches.

Abstract

We present a computationally simple and general purpose scheme for the detection of all salient object contours in real images. The scheme is inspired by the mechanism of surround influence that is exhibited in 80% of neurons in the primary visual cortex of primates. It is based on the observation that the local context of a contour significantly affects the global saliency of the contour. The proposed scheme consists of two steps: first find the edge response at all points in an image using gradient computation and in the second step modulate the edge response at a point by the response in its surround. In this paper, we present the results of implementing this scheme using a Sobel edge operator followed by a mask operation for the surround influence. The proposed scheme has been tested successfully on a large set of images. The performance of the proposed detector compares favourably both computationally and qualitatively, in comparison with another contour detector which is also based on surround influence. Hence, the proposed scheme can serve as a low cost preprocessing step for high level tasks such shape based recognition and image retrieval.

Abstract

Noncanonical base pairs in RNA have strong structural and functional implications but are currently not considered for secondary structure predictions. We present results of comparative ab initio studies of stabilities and interaction energies for the three standard and 24 selected unusual RNA base pairs reported in the literature. Hydrogen added models of isolated base pairs, with heavy atoms frozen in their away from equilibrium geometries, built from coordinates extracted from NDB, were geometry optimized using HF/6-31G** basis set, both before and after unfreezing the heavy atoms. Interaction energies, including BSSE and deformation energy corrections, were calculated, compared with respective single point MP2 energies, and correlated with occurrence frequencies and with types and geometries of hydrogen bonding interactions. Systems having two or more N-HO/N hydrogen bonds had reasonable interaction energies which correlated well with respective occurrence frequencies and highlighted the possibility of some of them playing important roles in improved secondary structure prediction methods. Several of the remaining base pairs with one N-HO/N and/or one C-HO/N interactions respectively, had poor interaction energies and negligible occurrences. High geometry variations on optimization of some of these were suggestive of their conformational switch like characteristics.

Abstract

Biological systems function with the aid of large number of proteins involved in complex network of chemical reactions. Interfering with the expression of proteins native or foreign leads to overall changes in the functioning of the biological systems . Manipulating the expression of proteins responsible for the manifestation of diseases has therefore been an interesting area for alternative efficient therapautic possibilities. Proteins are expressed from the correspondig genes involving the trancription and translation of messenger RNAs . Interfering with these RNA has emerged as an effective method of silencing unwanted genes. This project deals with designing effective methods for gene silencing through RNA Interference. RNA interference (RNAi) has emerged as an effective method of silencing gene expression in eukaryotic cells. This technology allows scientists to study the phenotypic effect of reducing expression levels of individual genes, and essentially negates the arduous task of producing single gene knockout or dominant negative expressing cell lines. RNAi has very quickly been utilized both in basic science and medical research. Among its many applications are the characterization and regulation of gene function, signaling pathway analysis and target validation. The potential therapeutic value of RNAi is indicated in recent studies showing that the RNAi pathway is functional in mice and is effective against a variety of infectious diseases including HIV and Hepatitis B. Since the early work describing the use of double-stranded RNA (dsRNA) for post-transcriptional gene silencing (PTGS) by Fire and Mello and the application of short interfering RNA (siRNA) to induce specific gene suppression in mammalian cells by Elbashir et al., there has been a flurry of RNAi-related developments and applications. The vast number of recent scientific journal articles, the scramble for intellectual property, and the formation of multiple biotechnology tools, kits, reagents and therapeutics companies all reflect the power and popularity of this technique. In 2002, Science voted RNAi scientific breakthrough of the year. This paper talks about a novel method of finding efficient siRNA targets on the mRNA sequences. It consists of two parts. In the first part program was written to select suitable targets based on the Elbashir's guidelines. In the second part involves prediction of secondary structure of siRNA antisense strands of the targets predicted in the first part and also the secondary structure of the mRNA sequence, and using these results to further filter the targets.

Abstract

The ultimate goal of molecular cell biology is to understand the physiology of living cells in terms of the information that is encoded in the genome of the cell and we would like to address the question how computational biology can help in achieving this goal. Genes coding for proteins is a very important pattern recognition problem in bioinformatics and this lecture focuses on some computational issues in gene identification.

Abstract

Timely agro-advisory service and efficient risk mitigation mechanisms can provide stableincome to the farmers. To provide timely agro-advisory service, e-Sagu systemhas beendeveloped by IIIT-Hyderabad and Media Lab Asia. It is an IT-based personalized agro-advisorysystemwhich is being developedto provide high-qualitypersonalized (farm-specific) agricultural expert adviceto each and everyfarmin a timely manner at thefarmer’s door-step without farmer asking a question. The results show that significantbenefits have flowntothe farmers interms of reducedinputs andenhancedyield.Regarding risk mitigation, the agri-insurance schemes around the worldwhether anadvanced or a developing nation like India are suffering due to catastrophic losses,covariate risks, asymmetric information and lack of quality data. These problems maypose a serious threat to the survival of the implementing agencies unless the there is aneffective monitoring systemin place, with enough man power and robust infrastructure.In this paper we explain how agri-insurance schemes can be effectively implemented ontop of eSagu system. The agri-insurance scheme can bevery cost-effective as it can useresources and processes of eSagu system. Also, as eSagu provides quality and reliabledata at farmlevel,the problemof asymmetric information can be effectively hedged.Most important, eSagualsoimproves the performance offormal risk mitigationmechanisms through farm-specific agro-advisory and monitoring system.

Abstract

To bridge the information gap between the agricultural expert and the farmer, International Institute of Information Technology (IIIT), Hyderabad has built the eSagu (“Sagu” means cultivation in Telugu language) system, which is an IT-based personalized agricultural extension system to improve agricultural productivity by disseminating a fresh expert agricultural advice to the farmers, both in a timely and personalized manner. In e-Sagu, the agricultural experts generate the expert advice based on the information about the crop situation received in the form of both text and digital photographs. In Kharif2 2004, a prototype was developed and implemented with 1051 farms. In the prototype, a team of agricultural experts stayed at IIIT, Hyderabad (India) and delivered 20,000 pieces of agricultural expert advice to 1051 cotton farms of three villages (Oorugonda, Gudeppad and Oglapur) in Atmakur Mandal of Warangal district in Andhra Pradesh state, India, by looking at digital photographs and other farm information supplied by some educated and experienced farmers (coordinators) in these villages. The pilot project was implemented successfully. In this article, an analysis of 51 registered farmers’ regarding compliance of advices and corresponding effect is reported. The following of expert advice on pest and disease management and IPM practices were analyzed by giving appropriate scores based on their effect on yield and input cost. The analysis showed that about fifty percent of farmers have followed the practices which increase yield and reduce input cost. A high positive correlation (r=0.46**) was observed between the compliance rate and the yields. Though some farmers obtained same yield as the farmers who have followed the advices, their input costs were significantly higher.