Image Segmentation and Object Detection for Automobile using OpenCV and CNN

Abstract

Image segmentation and object detection using CNN (Convolutional Neural Network) and OpenCV (Open-Source Computer Vision) is a popular research area in the field of computer vision and autonomous driving. This method employs deep learning techniques and image processing algorithms to detect and track objects in real-time from a video stream captured by a camera mounted on a vehicle. The main aim of this project is to develop an accurate and robust object detection system that can detect various types of objects such as vehicles, pedestrians, and bicycles on the road. The proposed system uses a pre-trained CNN model to detect objects and OpenCV for further image processing and filtering. The system is evaluated on a publicly available dataset and achieves high accuracy and detection rates for various objects. The results of this study show the potential of using deep learning and image processing algorithms for real-time object detection in autonomous vehicles and traffic control systems. This study examines the use of Convolutional Neural Network techniques that have been used for image segmentation and object detection in road traffic. The study explored the use of Gaussian filters in image pre-processing. The study also trained the model to detect road traffic objects and return output/feedback. The experimental result of the model was an accuracy of 96% across 26 classes and a recall of 92%. The study, therefore, recommends the use of object detection models in road traffic systems and autonomous vehicles.

Keywords: Artificial Intelligence (AI), Computer vision, Image segmentation, Object detection.

View PDF