Intelligent Highway Lane Center Identification from Surveillance Camera Video

Qiu, Mei; Chien, Stanley; Mulay, Anup Atul; Christopher, Lauren; Ding, Zhengming; Chen, Yaobin; Sturdevant, James R; Cox, Ed

doi:10.1109/itsc48978.2021.9564560

Cited by 5 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This technology operated on closed circuit television which required a separate monitor at TMCs for each camera, often without PTZ capability. In the early days, cameras were primarily used for qualitative traffic system monitoring, but as technology improved, the use of the cameras grew to include incident verification as well as research into shockwaves and other traffic flow characteristics [5]- [14]. Similar deployments in large urban areas occurred around the same time and further stimulated interest in deploying cameras to assist with freeway management activities.…”

Section: Freeway Cameramentioning

confidence: 99%

Methodology for Automatically Setting Camera View to Mile Marker for Traffic Incident Management

Mathew,

Malackowski,

Gartner

et al. 2023

JTTs

View full text Add to dashboard Cite

Traffic incident management (TIM) is a FHWA Every Day Counts initiative with the objective of reducing secondary crashes, improving travel reliability, and ensuring safety of responders. Agency roadside cameras play a critical role in TIM by helping dispatchers quickly identify the precise location of incidents when receiving reports from motorists with varying levels of spatial accuracy. Reconciling position reports that are often mile marker based, with cameras that operate in a Pan-Tilt-Zoom coordinate system relies on dispatchers having detailed knowledge for hundreds of cameras and perhaps some presets. During real-time incident dispatching, reducing the time it takes to identify the most relevant cameras and setting their view on the incident is an important opportunity to improve incident management dispatch times. This research develops a camera-to-mile marker mapping technique that automatically sets the camera view to a specified mile marker within the field-of-view of the camera. Over 350 traffic cameras along Indiana's 2250 directional miles of interstate were mapped to approximately 5000 discrete locations that correspond to approximately 780 directional miles (~35% of interstate) of camera coverage. This newly developed technique will allow operators to quickly identify the nearest camera and set them to the reported location. This research also identifies segments on the interstate system with limited or no camera coverage for decision makers to prioritize future capital investments. This paper concludes with brief discussion on future research to automate the mapping using LiDAR data and to set the cameras after automatically detecting the events using connected vehicle trajectory data.

show abstract

Section: Freeway Cameramentioning

confidence: 99%

Methodology for Automatically Setting Camera View to Mile Marker for Traffic Incident Management

Mathew,

Malackowski,

Gartner

et al. 2023

JTTs

View full text Add to dashboard Cite

show abstract

“…As regards lane keeping assistance (LKA), when the vehicle deviates from the lane, it will no longer be limited to sending an early warning message to the driver but actively control the steering wheel, correct the lateral position of the vehicle, correct the vehicle deviation back into the lane, and actively avoid lateral collision or other risks [ 19 – 21 ]. Lane centering control (LCC) can assist the driver to control the steering wheel, center the vehicle in the center of the lane, continuously control the vehicle to drive in the center of the lane, and cooperate with adaptive cruise [ 22 – 24 ]. As regards automatic lane change assistance, in the process of lane line recognition, we not only recognize the lane line of this lane but also add the lane line recognition of adjacent lanes.…”

Section: Introductionmentioning

confidence: 99%

An End-to-End Lane Detection Model with Attention and Residual Block

Wang

Xiao-ting

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Lane detection, as one of the most important core functions in the autonomous driving environment, is still an open problem. In particular, pursuing high accuracy in complex scenes, such as no line and multiple lane lines, is an urgent issue to be discussed and solved. In this paper, a novel end-to-end lane detection model combining the advantages of attention mechanism and residual block is proposed to address the problem. A residual block alleviates the possible gradient problem. An attention block can help the proposed model centralize on where to focus in the process of learning feature representation, which can make the model itself more sensitive to the feature representation of lane lines through convolutional operations. Additionally, the U-shaped structure with three downsampling operations preserves the image resolution and the original lane line information in the image to the greatest extent. The U-shaped structure can directly output the prediction results to eliminate many complex or unnecessary calculation processes. The experimental results on two public lane detection datasets show that the lane detection performance of the proposed model can achieve high accuracy, and the corresponding weight sizes are only 2.25 M. Finally, to further explain the effectiveness of the proposed model, the unavoidable troubles encountered in the experiment are discussed.

show abstract