Search

Your search keyword '"Krishna Mohan C"' showing total 34 results
Start Over Author "Krishna Mohan C"
34 results on '"Krishna Mohan C"'

12. View and Illumination Invariant Object Classification Based on 3D Color Histogram Using Convolutional Neural Networks

Author

Ijjina, Earnest Paul, Krishna Mohan, C., Hutchison, David, Series editor, Kanade, Takeo, Series editor, Kittler, Josef, Series editor, Kleinberg, Jon M., Series editor, Mattern, Friedemann, Series editor, Mitchell, John C., Series editor, Naor, Moni, Series editor, Pandu Rangan, C., Series editor, Steffen, Bernhard, Series editor, Terzopoulos, Demetri, Series editor, Tygar, Doug, Series editor, Weikum, Gerhard, Series editor, Jawahar, C.V., editor, and Shan, Shiguang, editor

Published
2015
Full Text
View/download PDF

15. Scene Classification in Remote Sensing Images using Dynamic Kernels

Author

Vishnu Chalavadi, Krishna Mohan C, and Rajeshreddy Datla

Subjects
Discriminative model, Computer science, Kernel (statistics), ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Cognitive neuroscience of visual object recognition, Spatial ecology, Spatial variability, Mixture model, Representation (mathematics), Aerial image, Remote sensing
Abstract

Classification of scenes across multi-sensor remote sensing images with different spatial, spectral, temporal resolutions involves identification of variable length spatial patterns of objects in a scene. So, it necessitates the use of local representations from different regions of a scene in order to comprehend the scene formation. In this paper, we propose a dynamic kernel based representation to handle the patterns of variable lengths in the scenes of remote sensing images. These kernels help to assimilate spatial variability captured using convolutional features in a Gaussian mixture model. The statistics of GMM facilitate the dynamic kernels in preserving the local spatial similarities while handling the changes in spatial content globally within the same scene. The efficacy of the proposed method using two variants of the dynamic kernels is demonstrated on three benchmark scene classification datasets, namely, UCM Land Use (21 classes), Aerial image dataset (30 classes), and NWPU-RESISC45 (45 classes). Our experiments show that the mean interval kernel is better discriminative as it makes use of first and second-order statistics of GMM.

Published
2021
Full Text
View/download PDF

18. Scene Classification in Remote Sensing Images using Dynamic Kernels

Author

Datla, R., Chalavadi, V., Krishna Mohan, C., Datla, R., Chalavadi, V., and Krishna Mohan, C.

Abstract

Classification of scenes across multi-sensor remote sensing images with different spatial, spectral, temporal resolutions involves identification of variable length spatial patterns of objects in a scene. So, it necessitates the use of local representations from different regions of a scene in order to comprehend the scene formation. In this paper, we propose a dynamic kernel based representation to handle the patterns of variable lengths in the scenes of remote sensing images. These kernels help to assimilate spatial variability captured using convolutional features in a Gaussian mixture model. The statistics of GMM facilitate the dynamic kernels in preserving the local spatial similarities while handling the changes in spatial content globally within the same scene. The efficacy of the proposed method using two variants of the dynamic kernels is demonstrated on three benchmark scene classification datasets, namely, UCM Land Use (21 classes), Aerial image dataset (30 classes), and NWPU-RESISC45 (45 classes). Our experiments show that the mean interval kernel is better discriminative as it makes use of first and second-order statistics of GMM. © 2021 IEEE.

Published
2021

19. Classification of human actions using pose-based features and stacked auto encoder

Author

Earnest Paul Ijjina and Krishna Mohan C

Subjects
0209 industrial biotechnology, Computer science, business.industry, Representation (systemics), 02 engineering and technology, Machine learning, computer.software_genre, Fuzzy logic, Autoencoder, 020901 industrial engineering & automation, Artificial Intelligence, Signal Processing, 0202 electrical engineering, electronic engineering, information engineering, Key (cryptography), 020201 artificial intelligence & image processing, Computer Vision and Pattern Recognition, Artificial intelligence, business, computer, Software
Abstract

Human action recognition using pose-based features is proposed.Use of key movements of actions for efficient input representation.Unsupervised pre-training using stacked auto encoder is exploited. Display Omitted In this paper, we propose a method for classification of human actions using pose based features. We demonstrate that statistical information of key movements of actions can be utilized in designing an efficient input representation, using fuzzy membership functions. The ability of stacked auto encoder to learn the underlying features of input data is exploited to recognize human actions. The efficacy of the proposed approach is demonstrated on CMU MOCAP and Berkeley MHAD datasets.

Published
2016
Full Text
View/download PDF

20. Vehicle Trajectory Prediction at Intersections using Interaction based Generative Adversarial Networks

Author

Roy, Debaditya, Ishizaka, Tetsuhiro, krishna Mohan, C, et al, ., Roy, Debaditya, Ishizaka, Tetsuhiro, krishna Mohan, C, and et al, .

Abstract

Vehicle trajectory prediction at intersections is both essential and challenging for autonomous vehicle navigation. This problem is aggravated when the traffic is predominantly composed of smaller vehicles that frequently disobey lane behavior as is the case in many developing countries. Existing macro approaches consider the trajectory prediction problem for lane-based traffic that cannot account when there is a high disparity in vehicle size and driving behavior among different vehicle types. Hence, we propose a vehicle trajectory prediction approach that models the interaction among different types of vehicles with vastly different driving styles. These interactions are encapsulated in the form of a social context embedded in a Generative Adversarial Network (GAN) to predict the trajectory of each vehicle at either a signalized or non-signalized intersection. The GAN model produces the most acceptable future trajectory among many choices that conform to past driving behavior as well as the trajectories of neighboring vehicles. We evaluate the proposed approach on aerial videos of intersections recorded in China where lane discipline is not followed by vehicles. The proposed GAN based approach demonstrates 6.4% relative improvement in predicting trajectories over state-of-the-art.

Published
2019

27. Medical Image Indexing and Retrieval Using Multiple Features

Author

Srinivas, M, krishna Mohan, C, Srinivas, M, and krishna Mohan, C

Abstract

Automatic medical image classification refers to assigning an medical image into a class, amonga number of image categories. Due to computational complexity, it is an important task in the content based image retrieval (CBIR). In this paper, we propose a method for classification and retrieval of medical images using multi-feature extraction method. Here, edge and patch based methods are used to extract feature vectors. We demonstrate that these features provide discriminative information useful for classification of medical images by considering eight categories of images, namely, hand, skull, chest, mammogram, chest side view, skull side view, knee and neck. Using multi-feature extraction method improves the accuracy of CBIR. In this paper, three different typesof similarity measures are used for comparing the query image with database images. The experimental results suggest that the proposed method has the ability to retrieve relevant images for a given input query image and provides goodretrieval performance than compared with the single feature extraction method.

Published
2013

32. Development of a novel 3-month drug releasing risperidone microspheres

Author

Bhanusree Yerragunta, Satheesh Jogala, Krishna Mohan Chinnala, and Jithan Aukunuru

Subjects
In vivo, microspheres, polycaprolactone blends, risperidone, risperdal consta, Pharmacy and materia medica, RS1-441, Analytical chemistry, QD71-142
Abstract

Objective: The purpose of this study was to develop an ideal microsphere formulation of risperidone that would prolong the drug release for 3 months in vivo and avoid the need for co-administration of oral tablets. Materials and Methods Polycaprolactones (PCL) were used as polymers to prepare microspheres. The research included screening and optimizing of suitable commercial polymers of variable molecular weights: PCL-14000, PCL-45000, PCL-80000 or the blends of these polymers to prepare microspheres with zero-order drug-releasing properties without the lag phase. In the present study, the sustained release risperidone microspheres were prepared by o/w emulsion solvent evaporation technique and the yield was determined. Microspheres were evaluated for their drug content and in vitro drug release. Microspheres prepared using a blend of PCL-45000 and PCL-80000 at a ratio of 1:1 resulted in the release of the drug in a time frame of 90 days, demonstrated zero-order drug release without lag time and burst release. This formulation was considered optimized formulation. Optimized formulation was characterized for solid state of the drug using differential scanning calorimetry, surface morphology using scanning electron microscopy and in vivo drug release in rats. Results: The surface of the optimized formulation was smooth, and the drug changed its physical form in the presence of blends of polymers and upon fabrication of microspheres. The optimized formulation also released the drug in vivo for a period of 90 days. Conclusions: From our study, it was concluded that these optimized microspheres showed great potential for a better depot preparation than the marketed Risperdal Consta™ and, therefore, could further improve patient compliance.

Published
2015
Full Text
View/download PDF

33. Development of SLN and NLC enriched hydrogels for transdermal delivery of nitrendipine: in vitro and in vivo characteristics.

Author

Bhaskar K, Krishna Mohan C, Lingam M, Jagan Mohan S, Venkateswarlu V, Madhusudan Rao Y, Bhaskar K, Anbu J, and Ravichandran V

Subjects
Administration, Cutaneous, Animals, Calcium Channel Blockers administration & dosage, Calcium Channel Blockers adverse effects, Calorimetry, Differential Scanning, Chemistry, Pharmaceutical, Delayed-Action Preparations, Drug Stability, Drug Storage, Hypertension drug therapy, Male, Microscopy, Electron, Scanning, Microscopy, Electron, Transmission, Nitrendipine administration & dosage, Nitrendipine adverse effects, Particle Size, Rats, Rats, Wistar, Skin Absorption, Triglycerides chemistry, Calcium Channel Blockers pharmacokinetics, Drug Carriers chemistry, Nanoparticles, Nitrendipine pharmacokinetics
Abstract

The purpose of this research was to investigate novel particulate carrier systems such as solid lipid nanoparticles (SLN) and nanostructured lipid carrier (NLC) for transdermal delivery of nitrendipine (NDP). For this investigation, four different gel-forming agents were selected for hydrogel preparation. Aqueous dispersions of lipid nanoparticles made from trimyristin (TM) were prepared by hot homogenization technique followed by sonication and then incorporated into the freshly prepared hydrogels. The particle size was analyzed by photon correlation spectroscopy (PCS) using Malvern zetasizer, which shows that for all the tested formulations, more than 50% of the particles were below 250 nm after 90 days of storage at room temperature. DSC analysis was performed to characterize the state of drug and lipid modification. Shape and surface morphology were determined by scanning electron microscope (SEM) and transmission electron microscope (TEM), which revealed fairly spherical shape of the formulations. The antihypertensive activity of the gels in comparison with that of oral NDP was studied using desoxy corticosterone acetate (DOCA)-induced hypertensive rats. It was observed that both carbopol SLN (A1) and carbopol NLC (B1) gels significantly controlled hypertension from the first hour (p < .05). The developed gels increased the efficacy of NDP for the therapy of hypertension. Both the SLN and NLC dispersions and the gels enriched with SLN and NLC possessed a sustained drug release over a period of 24 h, but the sustained effect was more pronounced with the SLN and the NLC gel formulations. Further, they were evaluated for zeta potential, entrapment efficiency, in vitro release, ex vivo permeation, and skin irritation studies.

Published
2009
Full Text
View/download PDF

34. Development of nitrendipine controlled release formulations based on SLN and NLC for topical delivery: in vitro and ex vivo characterization.

Author

Bhaskar K, Krishna Mohan C, Lingam M, Prabhakar Reddy V, Venkateswarlu V, and Madhusudan Rao Y

Subjects
Administration, Cutaneous, Animals, Calcium Channel Blockers administration & dosage, Chemistry, Pharmaceutical, Delayed-Action Preparations, Diffusion, Drug Stability, Drug Storage, In Vitro Techniques, Kinetics, Lipids chemistry, Male, Nitrendipine administration & dosage, Particle Size, Permeability, Rats, Rats, Wistar, Sonication, Calcium Channel Blockers pharmacokinetics, Nanoparticles, Nitrendipine pharmacokinetics, Skin Absorption
Abstract

The aim of the investigation is to develop solid lipid nanoparticles (SLN) and nano-structured lipid carrier (NLC) as carriers for topical delivery of nitrendipine (NDP). NDP-loaded SLN and NLC were prepared by hot homogenization technique followed by sonication, and they were characterized for particle size, zeta potential, entrapment efficiency, stability, and in vitro release profiles. Also the percutaneous permeation of NDPSLN A, NDPSLN B, and NDPNLC were investigated in abdominal rat skin using modified Franz diffusion cells. The steady state flux, permeation coefficient, and lag time of NDP were estimated over 24 h and compared with that of control (NDP solution). The particle size was analyzed by photon correlation spectroscopy (PCS) using Malvern zeta sizer, which shows that the NDPSLN A, NDPSLN B, and NDPNLC were in the range of 124-300 nm during 90 days of storage at room temperature. For all the tested formulations (NDPSLN A, NDPSLN B, and NDPNLC), the entrapment efficiency was higher than 75% after 90 days of storage. The cumulative percentage of drug release at 24 h was found to be 26.21, 30.81, and 37.52 for NDPSLN A, NDPSLN B, and NDPNLC, respectively. The results obtained from in vitro release profiles also indicated the use of these lipid nanoparticles as modified release formulations for lipophilic drug over a period of 24 h. The data obtained from in vitro release from NDPSLN A, NDPSLN B, and NDPNLC were fitted to various kinetic models. High correlation was obtained in Higuchi and Weibull model. The release pattern of drug is analyzed and found to follow Weibull and Higuchi equations. The permeation profiles were obtained for all formulations: NDPSLN A, NDPSLN B, and NDPNLC. Of all the three formulations, NDPNLC provided the greatest enhancement for NDP flux (21.485 +/- 2.82 microg/h/cm(2)), which was fourfold over control (4.881 +/- 0.96 microg/h/cm(2)). The flux obtained with NDPSLN B (16.983 +/- 2.91 microg/h/cm(2)) and NDPNLC (21.485 +/- 2.82 microg/h/cm(2)) meets the required flux (16.85 microg/h/cm(2)).

Published
2008
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results