My Projects
Street Litter Detection • Developed a street litter detector using YOLOv5 model based on the Euclidean approach.
• Trained on PennFudan and TACO datasets, achieving an F1 score of 89%.
• Deployed on real-time CCTVs for efficient litter detection. 
Self-Balancing Bike with LQR Controller • Developed and implemented a specialized LQR controller for a self-balancing bike with a single reaction wheel.
• Verified through simulations in V-rep software and successfully constructed the bike using Solidworks 2023.
• Focused on autonomous item delivery. 
IOT Integration on Shaker Machine • Developed a low-cost non-contact AC sensor to check the status of components of the shaker machine in the Product Reliability lab at TCE.
• Targeted at small-scale and manufacturing industries to prevent accidents caused by unattended machines.
• Automated monitoring and cloud-based monthly report storage were implemented. 
AI-assisted Waste Recycling System • Created three DL models achieving 87% accuracy in identifying road waste with CCTVs.
• Achieved a classification accuracy of 92% for categorizing waste into seven recycling/disposal categories.
• Classified plastic items with 85% accuracy into five subcategories based on shapes.
• Designed and deployed a user-friendly website and mobile app for e-commerce of recyclable items.
• Ongoing collaboration with ICCE for practical deployment. 
PondFishDet • Developed CLAHE-YOLOv8 and MSR-YOLOv8 algorithms to enhance underwater image analysis for accurate fish detection.
• Trained on real-time underwater images with varying lighting conditions.
• Achieved MAP scores of 0.964 for MSR-YOLOv8 and 0.970 for CLAHE-YOLOv8.
• Ranked Top 6 in the DePondFi ’23 Challenge at the National Conference on Computer Vision, Pattern Recognition, Image Processing, and Graphics-2023.