Collision time estimation in computer vision systems
By combining a monocular camera and Kalman filter in a computer vision system with object detection and ground plane estimation, the limitations of high-cost lidar in existing technologies are overcome, achieving low-cost and efficient collision time estimation, which is suitable for autonomous driving and advanced driver assistance systems.
CN121120760APending Publication Date: 2025-12-12TEXAS INSTRUMENTS INC
2 Cites 0 Cited by
Patent Information
- Application Number
- CN202511191045.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2016-10-19
- Filing Date
- 2017-10-18
- Publication Date
- 2025-12-12
AI Technical Summary
Technical Problem
Existing collision time estimation methods rely on high-cost lidar and radars with poor angular resolution, which limits their application in consumer vehicles.
Method used
Images are captured by a monocular camera in a computer vision system. The three-dimensional position and relative position of objects are calculated by combining object detection, structure of motion reconstruction (SfM) and ground plane estimation. The position and velocity are tracked using a Kalman filter, the time of collision (TTC) is estimated and the trajectory is checked.
Benefits of technology
A low-cost collision time estimation method is provided, applicable to autonomous driving and advanced driver assistance systems, improving the accuracy and efficiency of collision time estimation.
✦ Generated by Eureka AI based on patent content.
Abstract
The invention relates to collision time estimation in a computer vision system. The invention provides a method for estimating a time-to-collision (TTC) of a detected object in a computer vision system, the method comprising: determining (200) a three-dimensional (3D) position of a camera in the computer vision system; determining (202) a 3D position of the detected object based on a 2D position of the detected object in an image captured by the camera and an estimated ground plane corresponding to the image; calculating (204) a relative 3D position of the camera, a velocity of the relative 3D position, and an acceleration of the relative 3D position based on the 3D position of the camera and the 3D position of the detected object, wherein the relative 3D position of the camera is relative to the 3D position of the detected object; and calculating (206) the TTC of the detected object based on the relative 3D position, the velocity, and the acceleration.
Need to check novelty before this filing date? Find Prior Art