Vehicle Detection and Classification using Optimal Deep Learning on High-Resolution Remote Sensing Imagery for Urban Traffic Monitoring

Abstract

Remote sensing images (RSI), such as aerial or satellite images, produce a large-scale view of the Earth's surface, getting them used to track and monitor vehicles from several settings like border control, disaster response, and urban traffic surveillance. Vehicle detection and classification utilizing RSIs is a vital application of computer vision and image processing. It contains locating and identifying vehicles from the image. It is done using many approaches having object detection approaches, namely YOLO, Faster R-CNN, or SSD, that utilize deep learning (DL) for locating and identifying the image. Also, the vehicle classification from RSIs contains classifying them dependent upon their variety, like trucks, motorcycles, cars, or buses utilizing machine learning (ML) techniques. This article designed and develop an automated vehicle type detection and classification using a chaotic equilibrium optimization algorithm with deep learning (VDTC-CEOADL) on high resolution RSIs. The presented VDTC-CEOADL technique examines the high-quality RSIs for the accurate detection and classification of vehicles. The VDTC-CEOADL technique employs a YOLO-HR object detector with Residual Network as a backbone model to accomplish this. In addition, CEOA based hyperparameter optimizer is de-signed for the parameter tuning of the ResNet model. For the vehicle classification process, the VDTC-CEOADL technique exploits the attention based long short term memory (ALSTM) mod-el. The performance validation of the VDTC-CEOADL technique is validated on high resolution RSI dataset, and the results portrayed the supremacy of the VDTC-CEOADL technique in terms of different measures.