RAMIREDDY OBULAKONDA REDDY
@iare.ac.in
Associate Professor, CSE Department
Institute of Aeronautical Engineering
RESEARCH INTERESTS
Image Processing
Scopus Publications
Scopus Publications
M. Indrasena Reddy, M. Purushotham Reddy, R. Obulakonda Reddy, and A. Praveen
Springer Science and Business Media LLC
Palayanur Srinivasan, R Obulakonda Reddy, Kaki Anirudh Sai, and Jayanth Naidu
IEEE
India has been a huge market for secondhand market. The most popular second-hand items sold in India include used cars, furniture, electronic appliances, Mobile phones and etc. In general, a used car would cost lesser than a brand new car. Thus people prefer buying a used car. During the pandemic, the secondhand automobile industry in India has seen a sharp increase in popularity. Many people choose to purchase a used car rather than a brand-new one due to the difficulty of using public transit and the requirement for personal transportation. A used car was more cost-effective to purchase than a new one because of the economic uncertainty brought on by the pandemic and the increased prudence with which people approached their spending. Also, the pandemic’s effects on the supply chain and production delays made the used automobile market more alluring to customers. In this paper, the comparative analysis of multiple machine learning algorithms enabled us to identify the most effective method for determining the value of secondhand automobiles. Last but not least, providing an interactive web application to the owners of the cars where they can enter the details get an overview at the price. This study has processed the proposed dataset using popular Machine learning algorithms. Feature selection with Random Forest Regression yield the best performance with r$2_{-}$score = 0.9321 (accuracy of 93.2%) with MAE $\\sim$= 70213.9685. This study aims to increase the reliability of prediction of used car prices. In particular, this study concentrates on the approach of selecting features. The precise selection of features definitely showed us a significant edge over previous works.
K. Reddy Madhavi, Vineela Krishna Suri, V. Mahalakshmi, R. Obulakonda Reddy, and C. Sateesh kumar Reddy
Springer Nature Switzerland
M. Sudhakara, Y. Vijaya Shambhavi, R. Obulakonda Reddy, N. Badrinath, and K. Reddy Madhavi
Springer Nature Singapore
Sudhakara Malla, M. Janaki Meena, Obulakonda Reddy. R, V. Mahalakshmi, and Awatef Balobaid
Auricle Technologies, Pvt., Ltd.
Underwater Fish Species Recognition (UFSR) has attained significance because of evolving research in underwater life. Manual techniques to distinguish fish can be tricky and tedious. They might require enormous inspecting endeavours, but they can be costly. It results in limited data and a lack of human resources, which may cause incorrect object identification. Automating the fish species detection and recognition utilizing technology would assist sea life science to evolve further. UFSR in wild natural habitats is difficult because the images open natural habitat, complex background, and low luminance. Species Visualization can assist us with deep knowledge of the movements of the species underwater. Automation systems can help to classify the fish accurately and consistently. Image classification has been emerging research with the advancement of deep learning systems. The reason is that the convolutional neural networks (CNNs) don't require explicit feature extraction methods. The vast majority of the current object detection and recognition mechanisms are based on images in the outdoor environment. This paper mainly reviews the strategies proposed in the past years for underwater fish detection and classification. Further, the paper also presents the classification of three different underwater datasets using CNN with evaluation metrics.
K. Rangaswamy, D. Dhanya, B. Rupa Devi, Sateesh Kumar Reddy C, and R. Obulakonda Reddy
Auricle Technologies, Pvt., Ltd.
The high death rate and overall complexity of the cancer epidemic is a global health crisis. Progress in cancer prediction based on gene expression has increased in light of the speedy advancement using modern high-throughput sequencing methods and a wide range of machine learning techniques, bringing insights into efficient and precise treatment decision-making. Therefore, it is of significant interest to create machine learning systems that accurately identify cancer patients and healthy people. Although several classification systems have been applied to cancer prediction, no single strategy has proven superior. This research shows how to apply deep learning to an optimization method that uses numerous machine learning models. Statistical analysis has helped us choose informative genes, and we've been feeding those to five different categorization models. The results from the five different classifiers are ensembled in the next step using a deep learning technique. The three most common types of adenocarcinoma are those of the lungs, stomach, and breasts. The suggested deep learning-based inter-ensembles model was tested with deep learning-based algorithms on Carcinoma data. The results of the tests show that relative to using only one set of classifiers or the simple consensus algorithm, it improves the precision of cancer prognosis in every analyzed carcinoma dataset. The suggested deep learning-based inter-ensemble approach is demonstrated to be reliable and efficient for cancer diagnosis by entirely using diverse classifiers.
R. Obulakonda Reddy, Sandeep Kautish, V.Padmanabha Reddy, N. Sudhakar Yadav, Meznah M. Alanazi, and Ali Wagdy Mohamed
Hindawi Limited
An electrical device that transforms the electricity into the waves of radio and vice versa is termed the antenna. Its main deployment is in the transmitter and receiver of the antenna. While transmission, the transmitter of radio at the extremities of the antenna furnishes the electricity which oscillates at the frequency of radio wave and energy is released as current as em waves. Some of the voltage is formed from the em wave that is invaded at the point of receiving to amplify the receiver. This study focuses on the analysis of the satellite system to aid in mobile antenna tracking. It also examines the techniques for fuzzy control which make up traditional networks that are used. Initially, a basic idea of tracking loops with stabilized antennas was suggested in light of the requirement for the margin of phase and bandwidth. If the gain of the track is reduced due to changes in attributes and throughput, it will be reduced. In addition, fuzzy regulators and PID constituents are used to enhance the loop. The results indicate that the higher and lower antenna tracking gains within the loop were the best fit and the loop's fluctuations are reduced. A controller based on fuzzy logic can be most efficient due to its simplicity and robustness. It is also discovered that fuzzy logic controllers are evaluated by their behavior in relation. This paper presents an evaluation of the controllers in fuzzy logic, which is based on its integration with conventional controllers. There are three gains in PID's regulator PID and every gain can be used to control the variables of inputs and outcomes. The effects of the responses were analyzed and were compared. The commonality was discovered in the results according to the increase in time for II/6 and II/3 based on PID's regulator PID stability, it can be improved by this system, and there is a reduction in the duration of stability. Furthermore, the period of stability may be reduced through the fusion of PID and fuzzy. The effectiveness of the system could be enhanced by the implementation of the neural network. It is also possible to design the two types of control that could be used to control the proposed solid platform.
Seeja G., Obulakonda Reddy R., Korupalli V. Rajesh Kumar, S. S. L. C. H. Mounika, and Reddy Madhavi K.
IGI Global
The recent industrial scenarios project its advancements and developments with the intervention of integrated technologies including internet of things (IoT), robotics, and artificial intelligence (AI) technologies. Industrial 4.0 revolutions have broken the barriers of all restricted industrial boundaries with the act of those interdisciplinary concepts and have taken a keen part in industrial development. Incorporation of these advancements considerably helps in improving product efficiency and in reducing the production cost. Based on categories of production, industrial automation processes may vary. In this regard, robots are playing a vital role to automate the production process at various levels of industrial operations. The combination of IoT, robotics, and AI technologies enhances the industrial productivity towards getting the success rate. This chapter focuses on how robotic technology with IoT and AI methods enhances the limitations of various industrial applications.
Dr. Padmanabha Reddy V.
Institute of Advanced Scientific Research
Recognition and detection of an object in the watched scenes is a characteristic organic capacity. Animals and human being play out this easily in day by day life to move without crashes, to discover sustenance, dodge dangers, etc. Be that as it may, comparable PC techniques and calculations for scene examination are not all that direct, in spite of their exceptional advancement. Object detection is the process in which finding or recognizing cases of articles (for instance faces, mutts or structures) in computerized pictures or recordings. This is the fundamental task in computer. For detecting the instance of an object and to pictures having a place with an article classification object detection method usually used learning algorithm and extracted features. This paper proposed a method for moving object detection and vehicle detection.
Praveen Reddy J, Dr.R. Obulakonda Reddy, Dr.V. Padmanabha Reddy, and Elemasetty Uday Kiran
Institute of Advanced Scientific Research
B. Eswara Reddy, Ramireddy Obulakonda Reddy, and E. Keshava Reddy
Inderscience Publishers
V. Padmanabha Reddy, B V Ramnaresh Yadav, and R Obulakonda Reddy
Diva Enterprises Private Limited
In the domain of medical imaging, Computer Tomography (CT), Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET), Single Photon Emission Computed Tomography (SPECT) and rest of the other modes of medical images which will reflects human data from different views. We use medical images in radiography for diagnosis, clinical observations and in many other treatments. In this paper, we explored diverse medical imaging methods utilized for diagnosis of breast cancer, includes X-Ray mammography, MRI etc. we analyzed their adequacy, points of interest and weaknesses for distinguishing beginning time of breast cancer. Further we studied image fusion, a technique to expel the redundant information from the original medical images along with medical factors UIOQ, fusion factor, entropy etc. Be that as it may, from visual point of view, images that are fused gives better results than fuzzy systems. We further extended by investigating the medical tools to increase the speed and to enhance the study and analysis of medical images. The tools simulate the complete work, done by the human observations with numerous predefined components integrated in the software. However, protection software's or medical tools are challenging issue, since the tools provide content are routine in nature. We concluded our paper with few image encryption mechanisms along with their functionality.
Yerragudipadu SubbaRayudu, R. Obulakonda Reddy, and P. Anjaiah
IEEE
D2D communication in cellular networks is defined as direct communication between two mobile users without traversing the Base Station (BS) or core network. D2D communication is generally non-transparent to the cellular network and it can occur on the cellular frequencies (i.e., inband) or unlicensed spectrum (i.e., outband). The main aim of this paper is To consider a distributedcaching (Device-to-Device) D2Dnetwork. in which an Owner's file of interest is cached as several parts in the storage of other devices in the network. Here The Owner think that he needs to achieve all these file parts, the portions cached farther away naturally become the performance bottle-neck. This is due to the trust that Main interferers may be near to the receiver than the serving device. Using a simple stochastic geometry model, we concretely demonstrate that this bottleneck can be loosened if the Owners are mobile. Gains obtained from mobility (the ability to move or be moved freely and easily)are quantified in terms of coverage probability.
R. Obulakonda Reddy, B. Eswara Reddy, and E. Keshava Reddy
IEEE
Due to the lack of classification accuracy in pattern recognition, in this paper we propose a new algorithm for pattern analysis based on the symbolic pattern method. Proposed algorithm is constructed by using symbolic method and finite state automata model and used for classifying the textures based on the patterns. This algorithm performs symbolization of the data and portioning the texture images into two dimensional data. Features are extracted from the symbolic image and finite automata transition model. A probabilistic based classifier is designed to classify the texture images. The experimental study shows the better classification accuracy of the proposed system for the texture images for varied number of samples.