Rolling-Shutter NIR Imaging for Low-Light Driver Assistance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driver assistance systems in vehicles face challenges in low light conditions, particularly with lane marker detection, traffic sign recognition, and pedestrian detection, due to reduced sensitivity from infrared filters and color mosaic filters, which compromise performance in dark scenes and limit the simultaneous use of near-infrared illumination and color information.
Innovation Solution
A system with a camera and processor that uses a rolling shutter and near-infrared illuminator synchronized to the frame period, allowing for alternating exposure frames with and without near-infrared illumination, enabling concurrent performance of multiple driver assistance functions like pedestrian detection and automatic high-beam control while maintaining color information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If infrared filters and color mosaic filters are used in the camera, then color information can be captured, but sensitivity in low light conditions deteriorates
Solution Approach 1:
The camera sensor is divided into two distinct sensor arrays: a first sensor array without infrared or color filters that captures high-sensitivity monochrome images, and a second sensor array with infrared and color filters that captures color information. This segmentation allows each sensor to be optimized for its specific function, resolving the contradiction between color capture and low-light sensitivity.
2Measurement precision
If near-infrared illumination is used to improve detection range, then detection accuracy improves, but compatibility with other applications using color information deteriorates
Solution Approach 1:
The system segments the image data processing into two parallel paths: one path processes images from the first sensor array (without filters) enhanced by near-infrared illumination for high-precision detection of lane markers and traffic signs, while the other path processes images from the second sensor array (with filters) for color-based applications. This segmentation enables both near-infrared enhanced detection and color information usage simultaneously without interference.
Solution Approach 2:
The first sensor array without infrared or color filters serves multiple functions: it captures high-sensitivity monochrome images for near-infrared illumination enhancement, captures color information when needed, and provides the basis for both precision detection applications and color-based applications. This multi-functionality resolves the contradiction by making the sensor system adaptable to different application requirements.
3Device complexity
If a single sensor is used for both near-infrared detection and color information, then device complexity is reduced, but performance in both applications deteriorates
Solution Approach 1:
The camera is segmented into two sensor arrays with different filter configurations. The first sensor array lacks both infrared and color filters for maximum sensitivity and flexibility, while the second sensor array has both filters for dedicated color information capture. This segmentation maintains relatively simple device architecture while significantly improving performance in both near-infrared enhanced detection and color-based applications.
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
This solution enhances the detection range and accuracy of lane markers, traffic signs, and pedestrians in low light conditions, improving overall driver assistance system performance without sacrificing sensitivity or compatibility with other applications.
Implementation Method 1
A near infra-red illuminator is configured to provide a near infra-red illumination cone in the field of view of the camera
Implementation Method 2
The camera includes a rolling shutter configured to capture the image data during a frame period and to read the image data into multiple image frames
Implementation Method 3
A synchronization mechanism is configured to synchronize the illumination period to the frame period of the rolling shutter
Implementation Method 4
Images are received from an image sensor with a rolling shutter
Data Source
Figure 1a~1b
Figure 2~3
Figure 4a~4b
AI summary
A system (16) mountable in a motor vehicle includes a camera (12) and a processor (14) configured to receive image data from the camera (12). The camera (12) includes a rolling shutter (6) configured to capture the image data during a frame period and to scan and to read the image data into multiple image frames (204, 206). A near infra-red illuminator (4) may be configured to provide a near infra-red illumination cone in the field of view of the camera (12). The near infrared illumination oscillates with an illumination period. A synchronization mechanism may be configured to synchronize the illumination period to the frame period of the rolling shutter (6). The frame period may be selected so that the synchronization mechanism provides a spatial profile of the near infra-red illumination cone which may be substantially aligned vertically to a specific region, e.g. near the center of the image frame (204, 206).