Praveen Kumar Kollu | Work at Siddhartha Group of Instituions (original) (raw)

Papers by Praveen Kumar Kollu

2023 3rd Asian Conference on Innovation in Technology (ASIANCON)

International journal of emerging trends in engineering research, Aug 25, 2019

Protecting sensitive information over the internet requires specialized algorithms that can detec... more Protecting sensitive information over the internet requires specialized algorithms that can detect subtle abnormalities in the data. These abnormalities are harder to detect as most of the data contains redundant information that does not contribute to classification of normal and abnormal classes. Finding the most optimal features is still a difficult task. To optimize the model to focus on the most relevant features and to make the model train considerably faster, we are proposing an attention based recurrent neural network for intrusion detection in large scale networks. The model is optimized to leverage the power of GRU network while focusing on the most important features for classification. The latter is achieved by the attention vector that is introduced to the network. The efficiency of the network is determined using several measures such as accuracy, precision, recall and F1-score. These measures are compared among state-of-the-art classification algorithms to determine how our proposed model performs on par with the current approaches.

Turkish Journal of Computer and Mathematics Education (TURCOMAT), May 10, 2021

The need for Blockchain advancement, as well as the importance of its use, has fueled ongoing res... more The need for Blockchain advancement, as well as the importance of its use, has fueled ongoing research in a variety of theoretical and practical fields. Even as it is still in its early stages of development, the blockchain is being seen as a forward-thinking approach to modern technology issues such as decentralization, identification, confidence, character, data ownership, and information-driven decisions. The blockchain breakthrough offers major feedback when effectively searching for the optimal solution to storing and accessing cloud data. This essay examines the use of blockchain technologies to secure cloud computing. This research paper also presents a framework for secure storage of data in cloud computing environment. This framework makes use of smart contracts and access list for ensuring data security.

International journal of innovative technology and exploring engineering, Aug 30, 2019

Road network segmentation from high resolution satellite imagery have profound applications in re... more Road network segmentation from high resolution satellite imagery have profound applications in remote sensing. They facilitate for transportation, GPS navigation and digital cartography. Most recent advances in automatic road segmentation leverage the power of networks such as fully convolutional networks and encoder-decoder networks. The main disadvantage with these networks is that they contain deep architectures with large number of hidden layers to account for the lost spatial and localization features. This will add a significant computational overhead. It is also difficult to segment roads from other road-like features. In this paper, we propose a road segmentation architecture with an encoder and two path decoder modules. One path of the decode module approximates the coarse spatial features using upsampling network. The other path uses Atrous spatial pyramid pooling module to extract multi scale context information. Both the decoder paths are combined to fine tune the segmented road network. The experiments on the Massachusetts roads dataset show that our proposed model can produce precise segmentation results than other state-of-the-art models without being computationally expensive.

International journal of advanced trends in computer science and engineering, Aug 25, 2019

Computer networks are susceptible to a variety of security threats. With the ever growing number ... more Computer networks are susceptible to a variety of security threats. With the ever growing number of devices and people that are connecting to the network, it has become an utmost priority to defend these networks at a large scale. The constant changing nature of the networks and the increase in the type of attacks have made the traditional approaches to intrusion detection obsolete. In this paper, we are proposing a deeper recurrent neural network based approach for intrusion detection in large scale networks. The proposed model uses independent neurons in each layer to construct a deeper recurrent neural network. It helps in faster training and classification time as well as adaptability and scalability to dynamic environments. To evaluate our proposed model, CICIDS 2017 dataset was used to implement and compare against popular deep learning based approaches in network intrusion detection. The experiments have shown promising results that our proposed model can produce improved results over existing approaches.

SN Applied Sciences

Food instability has been linked to infertility, health issues, accelerated aging, incorrect insu... more Food instability has been linked to infertility, health issues, accelerated aging, incorrect insulin regulation, and more. Innovative approaches increased food availability and quality. Agriculture environment monitoring systems need IoT and machine learning. IoT sensors provide all necessary data for agriculture production forecast, fertilizer management, smart irrigation, crop monitoring, crop disease diagnosis, and pest control. Precision agriculture may boost crop yields by prescribing the right water-fertilizer-paste ratio. This article presents IOT based fertilizer recommendation system for Smart agriculture. This framework uses IoT devices and sensors to acquire agriculture-related data, and then machine learning is applied to suggest fertilizer in the correct quantity and at the appropriate time. The data acquisition phase collects input data, including soil temperature, moisture, humidity, regions' weather data, and crop details. Features are selected using the Sequenti...

2023 7th International Conference on Computing Methodologies and Communication (ICCMC)

2023 International Conference on Intelligent Data Communication Technologies and Internet of Things (IDCIoT), Jan 5, 2023

International Journal of Innovative Technology and Exploring Engineering, 2019

Road extraction from satellite images has several Applications such as geographic information sys... more Road extraction from satellite images has several Applications such as geographic information system (GIS). Having an accurate and up-to-date road network database will facilitate transportation, disaster management and GPS navigation. Most active field of research for automatic extraction of road network involves semantic segmentation using convolutional neural network (CNN). Although they can produce accurate results, typically the models give up performance for accuracy and vice-versa. In this paper, we are proposing architecture for semantic segmentation of road networks using Atrous Spatial Pyramid Pooling (ASPP). The network contains residual blocks for extracting low level features. Atrous convolutions with different dilation rates are taken and spatial pyramid pooling is performed on these features for extracting the spatial information. The low level features from residual blocks are added to the multi scale context information to produce the final segmentation image. Our p...

International Journal of Enterprise Network Management, 2018

The numbers of hacking and intrusion incidents are high due to the increasing use of internet ser... more The numbers of hacking and intrusion incidents are high due to the increasing use of internet services and computer application. Therefore, intrusion detection systems (IDS) are inevitable in today's scenario (Koruba et al., 2017). In this paper, an unsupervised technique based on hybrid clustering algorithms is used for Anomaly detection. Incremental support vector machine (ISVM) and C means (FCM) algorithms are applied to preprocess the data set and detect the anomalies respectively. Further, the processed data is fed to the DBSCAN algorithm for further detection of anomalies. The results of the detection system are communicated to the intrusion prevention system (IPS). The proposed hybrid algorithm is applied for KDD Cup 1999 dataset and Gure Kdd Cup data base (2008) and the results show high detection rates and low false positive alarms. Further, the proposed technique performs well with a real time data in detecting anomalies with enhanced true positive rate.

Journal of Healthcare Engineering, 2022

Internet of Things (IoT) is a successful area for many industries and academia domains, particula... more Internet of Things (IoT) is a successful area for many industries and academia domains, particularly healthcare is one of the application areas that uses IoT sensors and devices for monitoring. IoT transition replaces contemporary health services with scientific and socioeconomic viewpoints. Since the epidemic began, diverse scientific organizations have been making accelerated efforts to use a wide range of tools to tackle this global challenge and the founders of IoT analytics. Artificial intelligence (AI) plays a key role in measuring, assessing, and diagnosing the risk. It could be used to predict the number of alternate incidents, recovered instances, and casualties, also used for forecasting cases. Within the COVID-19 background, IoT technologies are used to minimize COVID-19 exposure to others by prenatal screening, patient monitoring, and postpatient incident response in specified procedures. In this study, the importance of IoT technology and artificial intelligence in COVI...

Int. J. Nat. Comput. Res., 2020

This article presents criminal bioinformatics approach which turned out to be fast, exact, and de... more This article presents criminal bioinformatics approach which turned out to be fast, exact, and definitive in the evaluation and the investigation of crude DNA profiling information. The most problematic scenario for mixture interpretation, however, is when the amount of DNA is limited for one or more of the sources in a mixture. The present study has examined the utility of legal bioinformatics application to Short Tandem Repeats (STR) information. The DNA profiling information is overseen and investigated on the grounds of the different loci display and changeability in various people. The authors have consolidated a similar general idea Inconstancy in STR areas can be utilized to recognize one DNA profile from another.

VOLUME-8 ISSUE-10, AUGUST 2019, REGULAR ISSUE, 2019

Tree Clustering from satellite images assists in ecological environmental protection. It also hel... more Tree Clustering from satellite images assists in ecological environmental protection. It also helps in managing green resources to provide sustainable development guidance. The automatic clustering of trees is a challenging task. Many models tend to give poor results when there is noise in the image. The aim is to propose a model for clustering of tree crown from panchromatic satellite image using image processing algorithms. In the proposed model we use Cartosat-2 satellite data and the image data is pre-processed to enhance the resulted image analyzed using segmentation models. The resulted image is trained using the clustering model which classifies the tree crowns from the panchromatic images. The proposed model can be able to classify tree crowns effectively from satellite imagery. The proposed model also calculates the tree crown height and width from satellite imagery.

Remote sensing plays a key role to monitor the earth resources. Remote sensing uses High Resoluti... more Remote sensing plays a key role to monitor the earth resources. Remote sensing uses High Resolution Satellites to capture and observe the various conditions of the earth like Land cover and Land use which provides the information regarding how much of land is covered by forest, wetland, water body and how much of land is used by people for rural development, urbanization and agricultural in digital images. Digital Image Processing is useful in decrypt satellite data which helps to know change detection and land cover classification. In this research, satellite data are used to investigate trees and identifying trees on the earth surface, where it is very difficult task to identify trees from high resolution satellite imagery. Digital Image Processing consists of various techniques like image enhancement, segmentation, feature extraction and classifying the extracted features. In this research, Cartosat2 images are used for detection and enumeration of trees. With utilization of digi...

International Journal of Emerging Trends in Engineering Research