Issuu

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 10 Issue: 04 | Apr 2023

p-ISSN: 2395-0072

www.irjet.net

Autonomous Driving Scene Parsing Neha Gupta1, Mohan Krishna Gupta2, Manan Mehrotra3, Prakhar Arora4, Rajul Bhatnagar5 1Assistant Professor of Computer Science and Engineering, MIT College, Uttar Pradesh, India 2Student of Computer Science and Engineering, MIT College, Uttar Pradesh, India 3Student of Computer Science and Engineering, MIT College, Uttar Pradesh, India 4Student of Computer Science and Engineering, MIT College, Uttar Pradesh, India 5Student of Computer Science and Engineering, MIT College, Uttar Pradesh, India ---------------------------------------------------------------------***--------------------------------------------------------------------Abstract - In this research paper, we give the semantic

segmentation of driving scenes in unconstrained conditions. The focus of previous approaches has been on constrained environments but we focus on the unconstrained environment of Indian roads. We have used Indian Driving Dataset (IDD[5]) which consists of 182 drive sequences on Indian roads. To perform semantic segmentation, we have used U-Net[6] model which is completely convolutional neural network, we have made some slight adjustments in the architecture for our purposes.

(a) Instance Segmentation

Examples of segmented images for an autonomous vehicle to aid in scene analysis include (a) an instance segmentation exemplar where various objects from nearly identical classes are segmented into various colours with their own boundary pixels, and (b) a semantic segmentation example where objects are illustrated with a single colour without any distinction.

Key Words: Deep learning, classification, convolutional neural networks, semantic segmentation, OpenCV.

1.INTRODUCTION utonomous Driving depends on information that has been processed from several sensors that are positioned above the car. These sensors enable the vehicle sense its surroundings, comprehend traffic scenes, and manage its motions, acting as its sense of hearing and sight. High-resolution cameras, radar, and Light Imaging Recognition and Ranging (LiDAR) are the main types of sensors used to classify objects using feature extraction and measure their distance from other objects using radio waves and illumination in order to create a three-dimensional (3D) image of the object. Environment. Several types of extra sensors, such as infrared, sonar, microradar, ultrasonic, and short-range sensors, have been fitted for autonomous vehicles to prevent collision with road obstacles. Similar to this, autonomous vehicles employ vision sensors to enable them to comprehend the visual components of their environment. Lane detection, traffic light and road sign analysis, pedestrian and vehicle detection, among other visual display and understanding problems, are all part of autonomous driving. By gathering this data, automated vehicle behaviours like lane changes, braking, and turning manoeuvres can be better and safer instructed.

Impact Factor value: 8.226

(b) Semantic Segmentation

Vision sensor data is arguably regarded as the most reliable source of information among those gathered for vehicle decision-making. As a result, this field of study has received much research and application in Intelligent Transportation Systems (ITS), primarily from the perspectives of machine learning and deep convolutional neural networks (CNN). By iteratively collecting model characteristics from the input image and optimizingly achieving better representations, deep CNNs are neural networks with a variety of functional layers for image processing. Similar techniques are utilised for scene analysis from vision data, where deep CNN is used to real-time photos, for example, to determine where a pedestrian is and how far away it is from an autonomous vehicle. In contrast to this streamlined overview of computer-generated landscapes, the complicated models that are now being proposed are capable of generating multiple labels like pedestrians and vehicles as well as the localization.

2. LITERATURE REVIEW To understand the development of autonomous driving research in recent years, it is necessary to organise a literature review to understand the various application areas by which autonomous driving has developed, as well as to recognise research gaps. Thus, the research process, approaches and findings of the literature review are introduced in the next sections.

ISO 9001:2008 Certified Journal

Page 942