17862821. METHOD AND APPARATUS WITH LANE GENERTION simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)
Contents
METHOD AND APPARATUS WITH LANE GENERTION
Organization Name
Inventor(s)
DAE HYUN Ji of Hwaseong-si (KR)
METHOD AND APPARATUS WITH LANE GENERTION - A simplified explanation of the abstract
This abstract first appeared for US patent application 17862821 titled 'METHOD AND APPARATUS WITH LANE GENERTION
Simplified Explanation
The abstract describes a method for generating lane information using a neural network. Here is a simplified explanation of the abstract:
- The method uses a neural network to generate a lane probability map based on an input image.
- Another neural network is used to generate lane feature information and depth feature information by applying the lane probability map.
- A third neural network is then used to generate depth distribution information based on the depth feature information.
- Spatial information is generated by combining the lane feature information and the depth distribution information.
- A fourth neural network is used to generate offset information, which represents the displacement between a lane position and a reference line.
- Finally, three-dimensional (3D) lane information is generated using the offset information.
Potential applications of this technology:
- Autonomous driving systems: The generated lane information can be used by autonomous vehicles to accurately detect and navigate within lanes.
- Advanced driver assistance systems (ADAS): The technology can be used to enhance ADAS features such as lane departure warning and lane keeping assist.
- Traffic management: The generated lane information can be used to monitor and analyze traffic patterns, aiding in traffic flow optimization and congestion management.
Problems solved by this technology:
- Accurate lane detection: The neural network-based approach improves the accuracy of lane detection compared to traditional methods, even in challenging conditions such as poor lighting or occlusions.
- Depth estimation: By incorporating depth feature information, the method can estimate the distance of the detected lanes from the vehicle, providing additional contextual information.
- 3D lane representation: The technology enables the generation of three-dimensional lane information, which can be useful for various applications such as path planning and obstacle avoidance.
Benefits of this technology:
- Improved safety: Accurate lane information can enhance the safety of autonomous vehicles and ADAS by enabling precise lane keeping and warning systems.
- Enhanced situational awareness: The depth information and 3D lane representation provide a more comprehensive understanding of the road environment, aiding in decision-making for autonomous systems.
- Robust performance: The use of neural networks allows the system to handle various challenging scenarios, making it more reliable and adaptable in real-world conditions.
Original Abstract Submitted
A method of generating lane information using a neural network includes generating a lane probability map based on an input image, generating lane feature information and depth feature information by applying the lane probability map to a second neural network, generating depth distribution information by applying the depth feature information to a third neural network, generating spatial information based on the lane feature information and the depth distribution information, generating offset information including a displacement between a position of a lane and a reference line by applying the spatial information to a fourth neural network, and generating three-dimensional (3D) lane information using the offset information.
