Driving Lane Detection on Smartphones using Deep Neural Networks

Abstract

Current smartphone-based navigation applications fail to provide lane-level information due to poor GPS accuracy. Detecting and tracking a vehicle’s lane position on the road assists in lane-level navigation. For instance, it would be important to know whether a vehicle is in the correct lane for safely making a turn, or whether the vehicle’s speed is compliant with a lane-specific speed limit. Recent efforts have used road network information and inertial sensors to estimate lane position. While inertial sensors can detect lane shifts over short windows, it would suffer from error accumulation over time. In this article, we present DeepLane, a system that leverages the back camera of a windshield-mounted smartphone to provide an accurate estimate of the vehicle’s current lane. We employ a deep learning--based technique to classify the vehicle’s lane position. DeepLane does not depend on any infrastructure support such as lane markings and works even when there are no lane markings, a characteristic of many roads in developing regions. We perform extensive evaluation of DeepLane on real-world datasets collected in developed and developing regions. DeepLane can detect a vehicle’s lane position with an accuracy of over 90%, and we have implemented DeepLane as an Android app.

Enhanced SCNN-Based Hybrid Spatial-Temporal Lane Detection Model for Intelligent Transportation Systems

“…These findings suggest that while s=(1,1) offers a slight improvement in performance metrics at the expense of increased inference time, s= (4,8) achieves a favorable balance between the two, and s= (8,16) demonstrates a decline in performance with no significant time efficiency gain. With s=1, convolution operations are computed layer by layer, which, due to separate computation and memory access for each convolutional layer, results in decreased GPU utilization efficiency and, consequently, a longer processing time.…”

“…This indicates that a sliding stride of s= (4,8) provides a more optimal balance between performance and temporal efficiency. Further incrementing the sliding stride to (8,16) resulted in a continued decrease in performance metrics, with No. 7 and No.…”

“…Perumal et al [15] tackled the issue of unstructured roads with their LaneScanNET, which showed promise in conditions such as shadows, fog, and dust; however, it was not specifically tailored for scenarios with uneven lighting. Bhandari et al [16] introduced the deep learning model Deeplane, which facilitated lane detection in the absence of lane markings, yet its detection accuracy and efficacy in complex scenarios like shadows require further enhancement. Oğuz et al [17] employed a one-dimensional deep learning technique to classify pixel intensity distributions for lane marking detection, overcoming low recognition rates in the presence of shadows and road imperfections, but it was not sufficiently adept at detecting curved paths.…”

A Real-Time Complex Road AI Perception Based on 5G-V2X for Smart City Security

“…On this basis, we propose a new method of marker detection based on a traditional detection algorithm and deep learning. First, we extract the overall road area by training the Mask R-CNN [11][12][13][14][15]. The identified road area is used as the constraint area, the lane mark is detected in the area, the obtained discrete lane-line featurepoint information is clustered by the least-squares method, and the lane lines are fitted in a different field of view using straight-line and curve-fitting models [16,17].…”

EGM: An Efficient Generative Model for Unrestricted Adversarial Examples

“…Artiicial intelligence (AI) algorithms have been widely used in our real life, such as machine translation [37], automatic driving [3,29], building smart Internet-of-Things (IoT) [18,24,25,47], and etc. Unfortunately, a lot of research shows that AI algorithms are weak especially when they face attacks from adversarial examples [7, 8, 41, 42, 48ś50, 52ś54], which prevents us from using AI algorithms to build secure systems.…”