Mono Camera and Sensor Fusion for Small Object Height Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sensor systems, such as LIDAR and stereo camera systems, are inadequate for reliably detecting small objects at long distances and determining their height, as they lack sufficient angular resolution and suffer from signal noise in depth maps.
Innovation Solution
A method and system utilizing a high-resolution mono camera and a second environment detection sensor, such as a stereo camera, radar, or LIDAR, to capture images and representations, allowing for simultaneous or sequential object detection and height determination through data fusion, with optional semantic segmentation using CNNs for improved object detection and shape analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LIDAR systems are used for distance determination, then distance accuracy is improved, but angular resolution is insufficient for detecting small objects at long distances
Solution Approach 1:
The system segments the detection task by using LIDAR for distance measurement and camera for angular resolution, combining their respective strengths to overcome individual limitations
Solution Approach 2:
The patent merges LIDAR and camera data through sensor fusion, integrating distance information from LIDAR with angular information from camera images to achieve both accurate distance determination and sufficient angular resolution for detecting small objects at long distances
2Difficulty of detecting and measuring
If stereo camera systems are used for 3D mapping, then lateral resolution is improved, but height determination of small objects becomes unreliable due to signal noise
Solution Approach 1:
The system uses the camera as an intermediary to provide angular resolution information that complements the stereo camera's lateral resolution, enabling reliable height determination by reducing dependence on noisy depth map signals alone
Solution Approach 2:
The patent changes the approach by combining angular resolution measurements with distance measurements from LIDAR, transforming the height determination process from relying solely on stereo depth maps to using a hybrid method that reduces signal noise impact
3Difficulty of detecting and measuring
If high-resolution camera systems are used to map small objects, then angular resolution is improved, but object detection becomes unreliable due to insufficient depth information
Solution Approach 1:
The system merges camera-based angular resolution with LIDAR-based distance measurement, combining the high angular resolution of the camera with the accurate depth information from LIDAR to achieve reliable object detection and height determination
Data Source
AI summary
The present disclosure relates to a method for the detection and determination of the height of objects by means of an environment detection system, including a first environment detection sensor and a second environment detection sensor of a vehicle, wherein at least one of the environment detection sensors is a mono camera. The method includes the following steps: capturing a mono image using the camera; capturing an environment representation using the second environment detection sensor; carrying out object detection in the mono image; carrying out object detection in the environment representation of the second environment detection sensor; measuring the distance from the object in the environment representation of the second environment detection sensor; and carrying out height determination of the detected object.