MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202641095063 A) filed by Adhiparasakthi Engineering College, Melmaruvathur on August 05, 2026, for Mlp–xgboost Hybrid Framework For Optimized Smart Parking Management And Violation Detection.
Inventors include Dr. C. Dhaya; Mr. G. Sekar; S. Mahalakshmi; Dr. R. Srivel; Ms. A. Jayanthi; and Ms. S. Subhashini.
The application for the patent was published on August 14, 2026, under issue no. 33/2026.
Abstract: Urban parking management and traffic violation detection represent critical challenges in modern smart city infrastructure, with traditional sensor-based and manual monitoring systems suffering from high costs, limited scalability, and reduced accuracy under real-world environmental conditions including illumination variation, occlusion, and high vehicle density. This invention proposes an intelligent MLP–XGBoost Hybrid Framework for Optimized Smart Parking Management and Violation Detection, integrating a Multilayer Perceptron (MLP) neural network and XGBoost gradient boosting classifier in a hybrid ensemble architecture, augmented by a Spatial Pyramid Pooling-based DenseNet (SPP-DenseNet) for multi-scale spatial feature extraction and optimized using the White Shark Optimization (WSO) algorithm for automatic hyperparameter tuning and accelerated convergence. The system applies a comprehensive preprocessing pipeline to structured parking records — encompassing data cleaning, categorical encoding, class imbal
Disclaimer: Curated by HT Syndication.
