Event-Predictive Imaging Control for Accident Moment Recording
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional imaging devices struggle to record important moments, such as vehicle accidents, due to limitations in memory capacity and frame rate settings, often missing critical events or erasing preceding images when trying to capture high-frame-rate footage.
Innovation Solution
An imaging device with an imaging unit, a predicting unit, and a changing unit that adjusts frame rate and resolution based on an accident prediction score calculated using deep neural networks from captured image data, allowing for timely and high-definition recording of critical events without external information delays.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of moving object
If a relatively low frame rate is set for capturing images, then the memory capacity can store longer-time images, but the drive recorder may fail to record an important moment at the time of the occurrence of the accident
Solution Approach 1:
The patent implements dynamic frame rate adjustment based on accident prediction scores. The frame rate is changed from a fixed low value to a variable value that increases when an accident is predicted, allowing the system to adapt to changing conditions and capture important moments while maintaining extended recording duration during normal operation
Solution Approach 2:
The system changes the frame rate parameter dynamically based on the accident prediction score. When the prediction score indicates a high probability of an accident, the frame rate is increased to ensure important moments are captured, while during normal operation the lower frame rate maintains extended recording duration
2Reliability
If a relatively high frame rate is set to record an important moment, then the drive recorder can capture the accident moment without missing it, but images before and after the occurrence of the accident may be erased and fail to be recorded due to memory capacity limits
Solution Approach 1:
The system dynamically adjusts the frame rate based on real-time accident prediction. During normal operation, the low frame rate extends recording duration. When an accident is predicted, the frame rate increases to capture the important moment, and the system manages memory to preserve relevant images before and after the event
Solution Approach 2:
The system performs preliminary actions by increasing the frame rate before the actual accident occurs, based on prediction scores calculated from current image data. This preliminary high-frame-rate capture ensures the important moment is recorded without missing it, while the system manages memory to retain sufficient preceding context
3Adaptability or versatility
If the frame rate is changed based on various pieces of information input from the outside, then the driving state can be monitored, but the system may fail to record an important moment due to information processing delays
Solution Approach 1:
The patent extracts and uses only the necessary features from image data for accident prediction, rather than processing all external information. This selective extraction enables faster processing and timely frame rate changes while maintaining the ability to monitor driving conditions
Solution Approach 2:
The system introduces an accident prediction score as an intermediary metric that synthesizes image data features. This intermediary enables rapid decision-making for frame rate adjustment without the delays associated with processing multiple external information sources
Data Source
AI summary
Provided are an imaging device, an image recording device, and an imaging method capable of recording an important moment. An imaging device includes an imaging unit and a changing unit. The imaging unit captures an image of the surroundings, and generates image data. The changing unit changes the frame rate of an image captured by the imaging unit in accordance with the degree of possibility that a predetermined event occurs, which is predicted based on the image data.