Multi-Sensor Image Processing for Occluded Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in determining safe trajectories and control strategies due to limitations in accurately detecting and processing images of objects in their environment, particularly in real-time, especially when objects are partially occluded or when multiple sources of information are not effectively integrated.
Innovation Solution
A method and system that utilize image-capture devices, RADAR, LIDAR, and other sensors to receive and process images of objects, integrating information about object characteristics to determine a control strategy for the vehicle, including using transforms to locate objects within images and account for occlusions, thereby enabling safe navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If image processing is performed in real-time to determine control strategies, then operational efficiency is improved, but measurement precision deteriorates due to partial occlusions and limited detection accuracy
Solution Approach 1:
The patent combines multiple information sources including image data from image-capture devices, RADAR data, LIDAR data, and other sensor inputs to create a comprehensive object detection system. This fusion of multiple data streams compensates for the limitations of individual sensors, particularly when objects are partially occluded, thereby maintaining measurement precision while enabling real-time processing for autonomous vehicle control strategies
2Measurement precision
If multiple sources of information are integrated to improve detection accuracy, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The control system is designed as a multi-functional platform that can process and integrate various types of data from different sensor modalities (cameras, RADAR, LIDAR, etc.) through a unified object detection module. This universal approach allows the same system architecture to handle multiple information sources without requiring separate processing pipelines for each sensor type, thereby managing device complexity while achieving high measurement precision through data fusion
3Measurement precision
If transforms are used to locate objects within images to account for occlusions, then measurement precision is improved, but loss of time increases due to additional processing steps
Solution Approach 1:
The system applies geometric transforms and occlusion analysis as preliminary processing steps early in the object detection pipeline, before final control strategy determination. By performing these computationally intensive operations upfront and using the results to guide subsequent faster processing stages, the system achieves accurate object location even with occlusions while minimizing overall processing time impact on real-time control decisions
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables autonomous vehicles to accurately determine control strategies based on real-time image processing, effectively navigating through complex environments by integrating multiple sources of information and accounting for occlusions, enhancing safety and operational efficiency.
Implementation Method 1
A method and system that utilize image-capture devices, RADAR, LIDAR, and other sensors to receive and process images of objects
Implementation Method 2
A method and system that utilize image-capture devices, RADAR, LIDAR, and other sensors to receive and process images of objects
Data Source
AI summary
Methods and systems for the use of detected objects for image processing are described. A computing device autonomously controlling a vehicle may receive images of the environment surrounding the vehicle from an image-capture device coupled to the vehicle. In order to process the images, the computing device may receive information indicating characteristics of objects in the images from one or more sources coupled to the vehicle. Examples of sources may include RADAR, LIDAR, a map, sensors, a global positioning system (GPS), or other cameras. The computing device may use the information indicating characteristics of the objects to process received images, including determining the approximate locations of objects within the images. Further, while processing the image, the computing device may use information from sources to determine portions of the image to focus upon that may allow the computing device to determine a control strategy based on portions of the image.