Vehicle Imaging Control With Dual Exposure for Early Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In-vehicle cameras face challenges with subject blur and decreased recognition rates due to prolonged storage periods, especially at low illuminance and high vehicle speeds, which hinder timely recognition of moving objects.
Innovation Solution
An imaging system with a control unit that manages two storage periods within a full-frame period, allowing for shorter and longer exposure times, enabling prompt recognition and reducing blur by outputting signals generated during the first storage period before the end of the second, and adjusting recognition frequency based on vehicle state information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If the storage period is lengthened to improve imaging in low illuminance, then the imaging quality in low light is improved, but subject blur occurs and recognition rate decreases
Solution Approach 1:
The imaging system divides the full-frame period into multiple storage periods with different durations (first storage period and second storage period). The first storage period has a shorter duration to capture moving objects with minimal blur for recognition, while the second storage period has a longer duration to capture sufficient light in low illuminance conditions. This segmentation allows the system to obtain multiple images with different exposure characteristics from the same full-frame period.
2Reliability
If the storage period is lengthened to suppress flicker from traffic lights, then flicker suppression is improved, but subject blur occurs and recognition rate decreases
Solution Approach 1:
The system segments the storage period into multiple intervals within one full-frame period. By performing recognition processing at different time points (after the first storage period and after the second storage period), the system can suppress flicker effects while maintaining the ability to recognize moving objects. The multi-stage storage approach allows flicker suppression without requiring a single prolonged storage period that would cause subject blur.
3Speed
If frame rate is increased to improve recognition timing, then response speed is improved, but the system cannot perform recognition before the end of the full-frame period
Solution Approach 1:
The system performs preliminary recognition processing after the first storage period (which is shorter than the second storage period) within the same full-frame period. This allows recognition to be performed earlier than waiting for the complete full-frame period to end, effectively reducing the recognition delay. The preliminary recognition result can be used for timely responses while the second storage period continues to accumulate light for low illuminance imaging.
4Speed
If vehicle speed increases to improve travel efficiency, then travel efficiency is improved, but subject blur occurs and recognition rate decreases
Solution Approach 1:
The system dynamically adjusts the recognition processing strategy based on vehicle speed. When the vehicle is moving at high speed, the system performs recognition processing after the first storage period (shorter duration) to capture moving objects with minimal relative motion blur. This dynamic adaptation of the recognition timing and storage period selection allows the system to maintain recognition accuracy while the vehicle travels at high speeds.
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 timely image recognition and reduced subject blur, improving recognition rates even at high speeds and low illuminance, allowing for earlier obstacle detection and improved image quality for display.
Implementation Method 1
photoelectric conversion devices that digitally count the number of photons incident on an avalanche photodiode (APD) and output a count value from a pixel as a photoelectrically converted digital signal
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An imaging system includes a plurality of pixels each including a sensor unit configured to output pulses corresponding to photons and a counter configured to count the number of pulses. The imaging system generates a signal on the basis of a difference between count values of the counter when a storage period starts and when the storage period ends, performs storage operations of a first storage period and a second storage period within one full-frame period, and performs a control process in which the first storage period is shorter than the second storage period and a signal generated in the first storage period is output between an end of the first storage period and an end of the second storage period. The imaging system performs a recognition process on the basis of the generated signal and changes a recognition frequency on the basis of driving state information about a driving state of a movable apparatus.