Vasudeva Acharya | National Institute of Technology Karnataka,Surathkal (original) (raw)

Papers by Vasudeva Acharya

Research paper thumbnail of Optimized Skeleton graph based CNN for Human Abnormal Detection in Video Streams

Sparklinglight Transactions on Artificial Intelligence and Quantum Computing

Human Action Recognition (HAR) is the process of understanding human actions and behavior. HAR ha... more Human Action Recognition (HAR) is the process of understanding human actions and behavior. HAR has a broad range of applications, and it has been focused on increasing the attention in various domain of computed vision. Abnormal detection from video stream is vigorous to guarantee the security in both outside spaces with the internal. Furthermore, the abnormal actions are really infrequent and rare, which makes the supervision process more challenging and difficult. In this research, skeleton graph-based Convolutional Neural Network (CNN) is devised for human abnormal activity detection. Here, the skeleton graph-based CNN (Skeleton graph_CNN) is devised based on the concept of classical convolution and skeleton graph generation. The human action recognition classifies the human actions into normal and abnormal class. The abnormal actions from the recognized outcome are detected with Skeleton graph_CNN, which provides the various actions of human as an output. The Skeleton graph_CNNg...

Research paper thumbnail of An Efficient Image Preprocessing Technique for Land Use Analysis by Using Remote Sensing Images

International Journal of Advanced Trends in Computer Science and Engineering, 2021

With the widespread availability of image manipulating software and with ease of use, anyone with... more With the widespread availability of image manipulating software and with ease of use, anyone with basic knowledge can modify the image without leaving any traces in the image. People believe in whatever they see in an image, but it is not necessary that the content has to be real. With the wide use of social media people with malicious intent share images that are modified and make you believe that they are real and by doing so they can influence people’s opinion which is a grave danger for society.In copy paste attack a part of the image is copied and pasted on to a different region of the same image, this pasted region can be detected by using the block-based method, but when the copied region is rotated orthogonally or if it is flipped before pasting on to the destination, such pasted regions are impossible to detect with the block-based approach.In this paper we are proposing a new technique to detect rotated or flipped pasted regions in an image and the performance of this new ...

Research paper thumbnail of An Efficient Image Preprocessing Technique for Land Use Analysis by Using Remote Sensing Images

International Journal of Advanced Trends in Computer Science and Engineering , 2021

The advancement in remote sensing technology expands the choice of selection of images for land u... more The advancement in remote sensing technology expands the choice of selection of images for land use mapping. Mapping, specific land use from remotely sensed images requires significant processing steps begins with image acquisition to classification using related algorithms. Useful properties of remote sensing images inherited with variety of errors requires some set of treatment in the form of preprocessing using various preprocessing techniques before the segmentation and classification of images. These preprocessing steps fine-tune the spatial and spectral feature of the data for future processing of the image. The residual characteristics of the image can be retained by calculating local mean and variance matching (LMVM) image fusion technique to retain the spectral information. The proposed study leverages the use of Just Noticeable Difference (JND) model to detect and map the edges of particular point of interest (POI) with reference to the luminance, contrast and structure of the remotely sensed image.

Research paper thumbnail of An algorithm for sum labeling of graphs

A simple undirected graph G is called a sum graph if there exists a labeling f of the vertices of... more A simple undirected graph G is called a sum graph if there exists a labeling f of the vertices of G into distinct positive integers such that any two distinct vertices u and v of G are adjacent if and only if there is a vertex w in G with label f (w) = f (u) + f (v). In this paper we give an algorithm for labeling a graph G of order n and size m when its adjacency matrix is given. We prove that The maximum integer in the sum labeling is bounded from above by 2 n 2 .3 n 2. Also we give an algorithm for optimal sum labeling of trees.

Research paper thumbnail of Optimized Skeleton graph based CNN for Human Abnormal Detection in Video Streams

Sparklinglight Transactions on Artificial Intelligence and Quantum Computing

Human Action Recognition (HAR) is the process of understanding human actions and behavior. HAR ha... more Human Action Recognition (HAR) is the process of understanding human actions and behavior. HAR has a broad range of applications, and it has been focused on increasing the attention in various domain of computed vision. Abnormal detection from video stream is vigorous to guarantee the security in both outside spaces with the internal. Furthermore, the abnormal actions are really infrequent and rare, which makes the supervision process more challenging and difficult. In this research, skeleton graph-based Convolutional Neural Network (CNN) is devised for human abnormal activity detection. Here, the skeleton graph-based CNN (Skeleton graph_CNN) is devised based on the concept of classical convolution and skeleton graph generation. The human action recognition classifies the human actions into normal and abnormal class. The abnormal actions from the recognized outcome are detected with Skeleton graph_CNN, which provides the various actions of human as an output. The Skeleton graph_CNNg...

Research paper thumbnail of An Efficient Image Preprocessing Technique for Land Use Analysis by Using Remote Sensing Images

International Journal of Advanced Trends in Computer Science and Engineering, 2021

With the widespread availability of image manipulating software and with ease of use, anyone with... more With the widespread availability of image manipulating software and with ease of use, anyone with basic knowledge can modify the image without leaving any traces in the image. People believe in whatever they see in an image, but it is not necessary that the content has to be real. With the wide use of social media people with malicious intent share images that are modified and make you believe that they are real and by doing so they can influence people’s opinion which is a grave danger for society.In copy paste attack a part of the image is copied and pasted on to a different region of the same image, this pasted region can be detected by using the block-based method, but when the copied region is rotated orthogonally or if it is flipped before pasting on to the destination, such pasted regions are impossible to detect with the block-based approach.In this paper we are proposing a new technique to detect rotated or flipped pasted regions in an image and the performance of this new ...

Research paper thumbnail of An Efficient Image Preprocessing Technique for Land Use Analysis by Using Remote Sensing Images

International Journal of Advanced Trends in Computer Science and Engineering , 2021

The advancement in remote sensing technology expands the choice of selection of images for land u... more The advancement in remote sensing technology expands the choice of selection of images for land use mapping. Mapping, specific land use from remotely sensed images requires significant processing steps begins with image acquisition to classification using related algorithms. Useful properties of remote sensing images inherited with variety of errors requires some set of treatment in the form of preprocessing using various preprocessing techniques before the segmentation and classification of images. These preprocessing steps fine-tune the spatial and spectral feature of the data for future processing of the image. The residual characteristics of the image can be retained by calculating local mean and variance matching (LMVM) image fusion technique to retain the spectral information. The proposed study leverages the use of Just Noticeable Difference (JND) model to detect and map the edges of particular point of interest (POI) with reference to the luminance, contrast and structure of the remotely sensed image.

Research paper thumbnail of An algorithm for sum labeling of graphs

A simple undirected graph G is called a sum graph if there exists a labeling f of the vertices of... more A simple undirected graph G is called a sum graph if there exists a labeling f of the vertices of G into distinct positive integers such that any two distinct vertices u and v of G are adjacent if and only if there is a vertex w in G with label f (w) = f (u) + f (v). In this paper we give an algorithm for labeling a graph G of order n and size m when its adjacency matrix is given. We prove that The maximum integer in the sum labeling is bounded from above by 2 n 2 .3 n 2. Also we give an algorithm for optimal sum labeling of trees.