AJAY KUMAR KUSHWAHA

AJAY KUMAR KUSHWAHA


Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 December 2023 in Analog Integrated Circuits and Signal Processing
Verified
Journal Article
Closed access
PDF file
SSSiddhesh Soyane
Siddhesh Soyane
AKAJAY KUMAR KUSHWAHA
AJAY KUMAR KUSHWAHA
DDDhiraj Manohar Dhane
Dhiraj Manohar Dhane

25 March 2023
Verified
Book Chapter
Closed access
PDF file
AKAJAY KUMAR KUSHWAHA
AJAY KUMAR KUSHWAHA
SKSupriya M. Khatavkar
Supriya M. Khatavkar
DBDhanashri Milind Biradar
Dhanashri Milind Biradar
PCPrashant A. Chougule
Prashant A. Chougule

28 January 2023
Verified
Book Chapter
Closed access
PDF file
AKAJAY KUMAR KUSHWAHA
AJAY KUMAR KUSHWAHA
AKAshok Kumar
Ashok Kumar
AMAshish Mishra
Ashish Mishra
ASAshutosh Singh
Ashutosh Singh

16 May 2025
Verified
Journal Article
Open access
PDF file
AKAJAY KUMAR KUSHWAHA
AJAY KUMAR KUSHWAHA
RPRajesh Prasad
Rajesh Prasad
JPJayashree Rajesh Prasad
Jayashree Rajesh Prasad
AJAnuja Jadhav
Anuja Jadhav
The rapid urbanization of cities worldwide presents a multitude of challenges, particularly in the domains of infrastructure management, disaster preparedness, environmental sustainability, and emergency response. Traditional city management and disaster response frameworks often depend on outdated methodologies, leading to inefficiencies in planning, forecasting, and execution. However, the advent of cutting-edge technologies such as Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) has opened new avenues for enhancing smart city development and disaster risk reduction. These technologies offer intelligent solutions by harnessing data-driven insights, predictive analytics, and real-time monitoring to improve decision-making processes. The analysis of historical temperature and precipitation data, along with Natural Earth datasets, was done using the Climatic Research Unit (CRU) and Natural Earth datasets to explore climate zones and future trends. We use geospatial visualization, clustering techniques such as K-Means and hierarchical clustering, and machine learning models to identify climate patterns across various regions. The findings reveal patterns in temperature and precipitation, similarities between regional climates, and the consequences of global climate changes. Furthermore, AI, ML, and DL applications for smart city development are discussed, demonstrating their contribution to optimizing urban infrastructure, resource management, and disaster preparedness. Data-driven climate analysis and emerging technologies are combined to support sustainable urban planning and enhance resilience against climate change.

5 January 2024
Verified
Book Chapter
Closed access
PDF file
PAPallavi Adke
Pallavi Adke
AKAJAY KUMAR KUSHWAHA
AJAY KUMAR KUSHWAHA
SKSupriya M. Khatavkar
Supriya M. Khatavkar
DSDipali Shende
Dipali Shende

26 September 2024
Verified
Journal Article
Open access
PDF file
TDTanuja Satish Dhope
Tanuja Satish Dhope
PDPallavi Deshpande
Pallavi Deshpande
AKAJAY KUMAR KUSHWAHA
AJAY KUMAR KUSHWAHA
DUDhananjay E. Upasani
Dhananjay E. Upasani
JGJyoti Gangane
Jyoti Gangane
In agriculture, plant leaf diseases cause significant productivity losses. Manual intervention in plant disease monitoring and control is time-consuming and involves a large human labour force. AI can address these difficulties by enabling the proactive diagnosis of plant diseases. Convolutional neural networks (CNNs) alone have demonstrated potential in automated plant disease diagnosis; but, when employed alone, their accuracy and resilience may be compromised. This study proposes an ensemble learning approach that combines four state-of-the-art CNN architectures: ResNet50, Efficient Net, Mobile net and VGG16. A comprehensive dataset of potato leaf images, categorized into healthy, early blight and late blight leaves, was used. Images were preprocessed and augmented to improve model generalization. Each model was trained separately, and their outputs were integrated using a weighted averaging mechanism to form the ensemble model. The weights for each model were optimized based on validation performance. The ensemble model significantly improved classification accuracy compared to individual models. The proposed method outperforms the other ensemble techniques with 5% improvement for precision and recall,10% reduction in response time,2% improvement in computational efficiency across all data test points.