Bangalore, India
Bangalore, India

Time filter

Source Type

Panigrahi N.,Center for and Robotics
Journal of the Indian Society of Remote Sensing | Year: 2015

Different classification techniques are being designed and under usage for classification of hyper spectral images. The usage of these classifiers differ for different type of hyper spectral data and application domain. The performance of these classifiers are influenced by feature preprocessing stage. In this research work we have investigated the impact of feature preprocessing using Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) on classification stage. The experiment is carried out using three sets of hyperspectral data. Classifications were carried out using three different classification techniques namely Maximum Likelihood Estimation (MLE), Constraint Energy Minimization (CEM) and Spectral Angle Mapper (SAM) on the preprocessed hyperspectral data. It is observed that the impact of PCA and LDA on the classification techniques are in two aspects (a) The preprocessing techniques facilitates to achieve high classification accuracy even with less number of training features and (b) Preprocessing expedites the classification process for large data sets. Also it can be concluded that PCA outperforms LDA in case of noisy data. © 2014, Indian Society of Remote Sensing.


Sharma A.M.,Center for and Robotics | Venkatesh K.S.,Indian Institute of Technology Kanpur | Mukerjee A.,Indian Institute of Technology Kanpur
ICIIP 2011 - Proceedings: 2011 International Conference on Image Information Processing | Year: 2011

Human pose estimation from monocular image sequences is attracting increasing attention, and 2D (image-based) as well as 3D (joint-motion based) approaches have been proposed. The former is computationally fast, and works also for less frequent poses, but reliability is low. The latter is computationally expensive owing to the high-dimensionality of the problem, despite attempts at dimensionality-reduction. We propose to impose temporal continuity constraints on the 2D approaches in order to improve reliability and obtain spatio-temporal descriptions of the action in the image plane. In the first stage, at each frame-level spatial localisation will be done by applying background subtraction, after that static image pose estimation is performed using CRF-based probabilistic assembly of parts, based on the approach by Ramanan [1]. Next, pose continuity is imposed via first-order Markovian constraints on the CRF search. This results in improved spatial accuracy as well as significantly reduced search. We present results from the Weizmann video dataset to demonstrate how such an approach can serve as an image-plane based alternative to full 3D modeling or manifold search. © 2011 IEEE.


Suri N.N.R.R.,Center for and Robotics | Murty M.N.,Indian Institute of Science | Athithan G.,Center for and Robotics
Proceedings of the 2012 12th International Conference on Hybrid Intelligent Systems, HIS 2012 | Year: 2012

The rapid growth in the field of data mining has lead to the development of various methods for outlier detection. Though detection of outliers has been well explored in the context of numerical data, dealing with categorical data is still evolving. In this paper, we propose a two-phase algorithm for detecting outliers in categorical data based on a novel definition of outliers. In the first phase, this algorithm explores a clustering of the given data, followed by the ranking phase for determining the set of most likely outliers. The proposed algorithm is expected to perform better as it can identify different types of outliers, employing two independent ranking schemes based on the attribute value frequencies and the inherent clustering structure in the given data. Unlike some existing methods, the computational complexity of this algorithm is not affected by the number of outliers to be detected. The efficacy of this algorithm is demonstrated through experiments on various public domain categorical data sets. © 2012 IEEE.


Faheema A.G.,Center for and Robotics | Rakshit S.,Center for and Robotics
2010 IEEE 2nd International Advance Computing Conference, IACC 2010 | Year: 2010

In this paper, we introduce an efficient method to substantially increase the recognition performance of object recognition by employing feature selection method using bag-of-visual-word representation. The proposed method generates visual vocabulary from a large set of images using visual vocabulary tree. Images are represented by a vector of weighted word frequencies. We have introduced on-line feature selection method, which for a given query image selects the relevant features from a large weighted word vector. The learned database image vectors are also reduced using the selected features. This will improve the classification accuracy and also reduce the overall computational complexity by dimensionality reduction of the classification problem. In addition, it will help us in discarding the irrelevant features, which if selected will deteriorate the classification results. We have demonstrated the efficiency our method on the Caltech dataset. ©2010 IEEE.


Paul M.V.V.,Center for and Robotics | Bhattacharjee R.,Center for and Robotics | Rajesh R.,Center for and Robotics
Proceedings - 2014 International Conference on Data Science and Engineering, ICDSE 2014 | Year: 2014

As the number of users on Internet is increasing day by day, Network traffic rate is going up exponentially. The traffic analysis capability also need to be scaled up correspondingly to handle this high traffic rates. The modern commodity platforms equipped with processors rich in number of cores and having hyper threading technology has enabled emergence of software based solutions rather than dedicated platforms for traffic analysis. But the general purpose network stacks provided by operating systems focus more on compatibility and hence are not able to fully exploit the hardware capabilities of these platforms. Lot of work is reported on customizing the stack for packet capture and achieving line rate (10 Gbps) performance on a single Network Interface Card (NIC). Some work is also reported on use of multiple 10GbE NICs for handling traffic beyond 10 Gbps. This paper brings out through experimental setup and observations the issues involved in linear scaling of performance with additional NICs on commodity Non Uniform Memory Access (NUMA) servers. It shows that linear scaling requires optimum configurations of compute, memory and network resources especially affinity of NICs to NUMA nodes. Using these configurations we are able to achieve linear scaling of traffic analysis capability with additional 10 GbE NICs. We report scaled performance upto 40 Gbps with the use of four NIC cards. © 2014 IEEE.


Suri N.N.R.R.,Center for and Robotics | Murty M.N.,Indian Institute of Science | Athithan G.,Center for and Robotics
Proceedings of the 2012 12th International Conference on Hybrid Intelligent Systems, HIS 2012 | Year: 2012

Outlier detection in high dimensional categorical data has been a problem of much interest due to the extensive use of qualitative features for describing the data across various application areas. Though there exist various established methods for dealing with the dimensionality aspect through feature selection on numerical data, the categorical domain is actively being explored. As outlier detection is generally considered as an unsupervised learning problem due to lack of knowledge about the nature of various types of outliers, the related feature selection task also needs to be handled in a similar manner. This motivates the need to develop an unsupervised feature selection algorithm for efficient detection of outliers in categorical data. Addressing this aspect, we propose a novel feature selection algorithm based on the mutual information measure and the entropy computation. The redundancy among the features is characterized using the mutual information measure for identifying a suitable feature subset with less redundancy. The performance of the proposed algorithm in comparison with the information gain based feature selection shows its effectiveness for outlier detection. The efficacy of the proposed algorithm is demonstrated on various high-dimensional benchmark data sets employing two existing outlier detection methods. © 2012 IEEE.


Singh S.,Center for and Robotics | Krishna K.M.,Indian Institute of Technology Hyderabad
Proceedings of the IEEE International Conference on Control Applications | Year: 2013

In this paper we develop an algorithm to generate gait sequences to negotiate a discontinuous terrain for a hybrid 4-wheeled legged robot. The gait sequence comprises two main steps - normal force redistribution and hybrid position-force control. The robot climbs the discontinuity one leg at a time. This requires that the entire load of the robot is taken up by the other three legs so that the leg climbing the discontinuity is free. For this purpose a load redistribution methodology is used which makes the center of gravity of chassis coincide with the desired center of pressure (CoP). Subsequently the free leg moves in hybrid position and force control to climb the discontinuity. Force sensing ensures constant contact with the terrain and detection of stand and end of the discontinuity without using any perception sensor. The methodology is validated using multi-body dynamic simulation. © 2013 IEEE.


Ravi V.C.,Center for and Robotics | Rakshit S.,Center for and Robotics | Ghosal A.,Indian Institute of Science
Journal of Mechanisms and Robotics | Year: 2010

Hyper-redundant robots are characterized by the presence of a large number of actuated joints, a lot more than the number required to perform a given task. These robots have been proposed and used for many applications involving avoiding obstacles or, in general, to provide enhanced dexterity in performing tasks. Making effective use of the extra degrees-of-freedom or resolution of redundancy has been an extensive topic of research and several methods have been proposed in literature. In this paper, we compare three known methods and show that an algorithm based on a classical curve, called the tractrix, leads to a more "natural" motion of the hyper-redundant robot with the displacements diminishing from the end-effector to the fixed base. In addition, since the actuators nearer the base "see" a greater inertia due to the links farther away, smaller motion of the actuators nearer the base results in better motion of the end-effector as compared with other two approaches. We present simulation and experimental results performed on a prototype eight-link planar hyper-redundant manipulator. © 2010 by ASME.


Singh S.,Center for and Robotics | Jadhav B.D.,Center for and Robotics | Krishna K.M.,Indian Institute of Technology Hyderabad
Proceedings - IEEE International Conference on Robotics and Automation | Year: 2014

In this paper, we present a posture control scheme for step climbing by an in-house developed three-segmented tracked robot, miniUGV. The posture control scheme results in minimum torque at the actuated joints of the segments. Non-linear optimization is carried out offline for progressively decreasing distance of the robot from the step with torque minimization as objective function and force balance, motor torque limits, slippage avoidance and interference avoidance constraints. The resulting angles of the joints are fitted to a third degree polynomial as a function of the robot distance from the step and the step height. It is shown that a single set of polynomial functions is sufficient for climbing steps of all permissible heights and angles of attack of the front segment. The methodology has been verified through simulation followed by implementation on the real robot. As a consequence of this optimization we find that the average current reduced by more than thirty percent, reducing power consumption and confirming the efficacy of the optimization framework. © 2014 IEEE.


Suri N.N.R.R.,Center for and Robotics | Murty M.N.,Indian Institute of Science | Athithan G.,Scientific Analysis Group
Natural Computing | Year: 2015

Outlier detection is an important data mining task with many contemporary applications. Clustering based methods for outlier detection try to identify the data objects that deviate from the normal data. However, the uncertainty regarding the cluster membership of an outlier object has to be handled appropriately during the clustering process. Additionally, carrying out the clustering process on data described using categorical attributes is challenging, due to the difficulty in defining requisite methods and measures dealing with such data. Addressing these issues, a novel algorithm for clustering categorical data aimed at outlier detection is proposed here by modifying the standard (Formula presented.)-modes algorithm. The uncertainty regarding the clustering process is addressed by considering a soft computing approach based on rough sets. Accordingly, the modified clustering algorithm incorporates the lower and upper approximation properties of rough sets. The efficacy of the proposed rough (Formula presented.)-modes clustering algorithm for outlier detection is demonstrated using various benchmark categorical data sets. © 2015 Springer Science+Business Media Dordrecht

Loading Center for and Robotics collaborators
Loading Center for and Robotics collaborators