Vehicular AI Object Detection for Selective Image Recording
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicular electronic devices face challenges in accurately recording and storing relevant information during vehicle operations, leading to unnecessary data storage and overhead due to continuous recording, and fail to effectively distinguish irrelevant objects, resulting in inefficient processing and storage.
Innovation Solution
A vehicular electronic device and method utilizing deep learning to detect objects in bounding box and skeleton formats, determining event occurrence based on predefined conditions, and selectively recording and transmitting images to minimize unnecessary data storage and processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If continuous recording is performed to ensure no relevant information is missed, then reliability of information capture is improved, but data storage volume and processing overhead increase significantly
Solution Approach 1:
The system performs preliminary object detection and classification before recording decisions are made. By pre-identifying relevant objects and their types using deep learning models, the system can determine in advance whether recording is necessary, avoiding unnecessary continuous recording and reducing data storage volume while maintaining information capture reliability.
Solution Approach 2:
The system uses feedback from object detection results to dynamically control the recording process. Detection outcomes feed into a decision-making mechanism that adjusts recording activation based on identified object types and potential interference with vehicle operations, creating a closed-loop system that optimizes both reliability and storage efficiency.
2Measurement precision
If continuous recording is performed to capture all potential events, then measurement completeness is improved, but processing time and computational resources increase
Solution Approach 1:
The system extracts and focuses only on relevant objects and events from the continuous video stream using deep learning-based object detection. By identifying and isolating specific object types that may interfere with vehicle operations, the system processes only necessary portions of the data, reducing overall processing time while maintaining event detection completeness.
Solution Approach 2:
The system applies partial action by selectively activating recording only when specific object types are detected that pose potential interference risks. Instead of processing all video data uniformly, the system applies detection and recording resources partially and selectively to relevant segments, reducing computational overhead while ensuring complete capture of critical events.
3Loss of information
If all detected objects are recorded to ensure comprehensive monitoring, then information completeness is improved, but storage efficiency deteriorates due to irrelevant data
Solution Approach 1:
The system applies local quality by differentiating between relevant and irrelevant objects based on their types and characteristics. Different object categories receive different treatment in terms of recording activation, with only those objects that may interfere with vehicle operations triggering recording. This selective approach maintains information completeness for critical events while improving storage efficiency by excluding irrelevant data.
4Measurement precision
If deep learning-based object detection is implemented to distinguish relevant objects, then detection accuracy is improved, but device complexity increases
Solution Approach 1:
The system segments the object detection task into distinct categories and types, using specialized deep learning models for different object classes. By dividing the detection problem into manageable segments rather than attempting to detect all objects uniformly, the system achieves high detection accuracy for relevant objects while managing computational complexity through modular model architecture.
Data Source
AI summary
A method comprises acquiring an image via image capturing; detecting an object in the image in a specific format based a deep learning; performing post-processing which determines whether a status of the detected object satisfies a predetermined condition; detecting, based on the post-processing, whether an event for activating recoding occurs; and starting recoding the image. An electronic device comprises a camera unit installed in a vehicle and configured to acquire an image of surroundings of the vehicle; a processor configured to detect an object in the image in a specific format based on deep learning, perform post-processing which determines whether a status of the detected object satisfies a predetermined condition, detect based on the post-processing whether an event for activating recoding occurs, and start recoding the image; and a storage unit configured to store the recorded image.