Sub-scene Target Detection and Recognition Using Deep Learning Convolution Neural Networks

Merugu, Suresh; Jain, Kamal; Mittal, Anshul; Raman, Balasubramanian

doi:10.1007/978-981-15-1420-3_119

Cited by 20 publications

(8 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Ministry of Health and Family Welfare, Government of India conducted a survey for more than 85,000 people "The National Blindness & Visual Impairment Survey India 2015-2019" recently. The overall estimated prevalence of the Blindness in age 50 or more is 1.99% and the overall estimated prevalence of Visually Impaired in age 50 or more is 13.76%was observed in the survey [12]. The above facts show that there is a significant number of individuals facing vision problems.…”

Section: Introductionmentioning

confidence: 69%

A Review of Some Assistive Tools and their Limitations for Visually Impaired

Merugu¹,

Ghinea²,

Shaik³

et al. 2022

helix

Self Cite

View full text Add to dashboard Cite

World report on Vision by World Health Organization reported that there are at least 2.2 Billion people are Visually Impaired. "The National Blindness & Visual Impairment Survey India 2015-2019" survey report by Government of India reported that the overall estimated prevalence of the Blindness in age 50 or more is 1.99% and the overall estimated prevalence of Visually Impaired in age 50 or more is 13.76%. The above numbers urge researchers to develop advanced assistive tools for Blind and Visually Impaired. There have been several assistive tools already designed to assist and enhance the quality of life to Blind and Visually Impaired. But most of them are limited and confined to specific problems. In this paper, a systematic review of some of the assistive tools for Blind and Visually Impaired is done. The detailed review of existing assistive tools reveals that the factors like user-friendly, ensuring safety, independent mobility, multiple features and user-centric design need priority consideration while designing such tools.

show abstract

Section: Introductionmentioning

confidence: 69%

A Review of Some Assistive Tools and their Limitations for Visually Impaired

Merugu¹,

Ghinea²,

Shaik³

et al. 2022

helix

Self Cite

View full text Add to dashboard Cite

show abstract

“…Haseena Kuttomparambil Abdul khader [20] used a metaheuristic hybrid algorithm to present a Type-2 fuzzy fractional-order PID (T2FFOPID) centered PSS, which promoted the power device's electromechanical oscillation damping performance to enrich the dynamic steadiness. In opposition to a huge variety of device fuctuations found in the MG, the massive bandwidth, reminiscence impact, along with fat phase contribution in the frequency reaction of FOPID were formulated to craft the controller and were executed nicely.…”

Section: Related Workmentioning

confidence: 99%

An FRLQG Controller-Based Small-Signal Stability Enhancement of Hybrid Microgrid Using the BCSSO Algorithm

Ensermu

Santhi

Merugu

et al. 2023

Journal of Electrical and Computer Engineering

View full text Add to dashboard Cite

The development of a network termed microgrid (MG) has been motivated owing to augmentation in renewable energy source (RES) infiltration along with the utilization of enhanced power electronic technologies. Recently, more popularity has been gained by the hybrid MG (HMG). Maintaining the power system’s (PS) small-signal stability (SSS) is highly complicated during the energy enhancement of RES. The enhancement of the SSS has been focused on by numerous existing methodologies; however, the optimal solution was not obtained by those methodologies. A new controller with the assistance of bell-curved squirrel search optimization (BCSSO) is proposed to address the aforementioned issue. Initially, for PSs such as photovoltaic (PV), wind turbines, along with fuel cells, a mathematical model is ascertained. Then, in this, the converter design has been developed. The PV’s maximum power flow is recognized by maximum power point tracking (MPPT) in the bidirectional switched buck-boost converter (BSBBC), which is utilized in this research, and by utilizing the fuzzy ruled linear quadratic Gaussian (FRLQG), the converters are controlled to assure safe operation along with soft dynamics. By employing the BCSSO, the parameters are modified in this controller which in turn ameliorates the SSS. The experiential evaluation of the proposed system’s performance is analogized with the existing methodologies. Consequently, the outcomes confirmed that a better performance was attained by the proposed methodology than the prevailing works.

show abstract

“…But, it results in high false alarm rates. ANN [13,21] provides a low false alarm rate and high detection rates. Still, the major limitations are the high cost of signature matching and network packet overload.…”

Section: Problem Statementmentioning

confidence: 99%

Designing Framework for Intrusion Detection in IoT Based on Spotted Hyena-Based ANN

Bathula

Muhuri

Merugu

et al. 2021

Lecture Notes in Electrical Engineering

Self Cite

View full text Add to dashboard Cite

Internet of Things (IoT) is the platform for resource sharing amidst the computing platforms, mainly memory, power, large data, and storage. Securing the data in the IoT has become a prime concern as the hackers can access through the invaluable information from the IoT database. Once the attacks are incurred on the IoT environment, the loss of data is inevitable, and it can have a major effect on the progress of the IoT platform. In this paper, an Intrusion Detection System (IDS) with improved artificial intelligence is proposed. This paper makes use of standard benchmark datasets from diverse sources for performing the experiment. A wellperforming machine learning algorithm called Artificial Neural Network (ANN) is developed with improved network architecture. The usage of a renowned metaheuristic algorithm called Spotted Hyena Optimization (SHO) is used for selecting the optimal hidden neurons for ANN. The main objective of the improved training is to reduce the error difference between the target and the measured outputs to enhance the detection accuracy. Finally, the experimental outcomes and simulations prove the stability and robustness of the proposed model in terms of a variety of performance metrics over other machine learning models.

show abstract

Sub-scene Target Detection and Recognition Using Deep Learning Convolution Neural Networks

Cited by 20 publications

References 21 publications

A Review of Some Assistive Tools and their Limitations for Visually Impaired

A Review of Some Assistive Tools and their Limitations for Visually Impaired

An FRLQG Controller-Based Small-Signal Stability Enhancement of Hybrid Microgrid Using the BCSSO Algorithm

Designing Framework for Intrusion Detection in IoT Based on Spotted Hyena-Based ANN

Contact Info

Product

Resources

About