Video Monitoring Feed Switching Based on Observer Attention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video monitoring systems face challenges in automatically switching between camera feeds to match the observer's intention, leading to increased workload and potential missed abnormalities, especially with a large number of cameras.
Innovation Solution
A video monitoring apparatus that acquires images from multiple cameras, estimates user attention degrees, learns to prioritize selected images, and automatically switches displays based on these attention degrees, reducing observer workload and improving abnormality detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If automatic abnormality detection is performed using pre-labeled normal and abnormal data, then abnormality detection capability is improved, but the detection result may not match the observer's intention leading to missed abnormalities
Solution Approach 1:
The system incorporates feedback loops where observer corrections to automatic detection results are used to retrain and improve the detection model. The observer's manual selections and corrections provide feedback signals that adjust the weightings and parameters of the abnormality detection algorithm, enabling the system to learn and adapt to specific observer intentions and requirements over time.
2Reliability
If abnormality degrees are accumulated in a database and searched, then comprehensive abnormality analysis is improved, but detection time increases significantly with large data amounts
Solution Approach 1:
The system segments the abnormality detection process into multiple stages: real-time feature extraction from video streams, intermediate scoring based on multiple abnormality types, and final ranking. This segmentation allows parallel processing of different feature extraction tasks and enables the system to provide preliminary results quickly while continuing to process comprehensive analysis in the background, reducing overall detection time without sacrificing comprehensiveness.
Solution Approach 2:
The system performs preliminary actions by pre-calculating and storing baseline feature representations of normal video content, and by pre-defining multiple abnormality detection models for different types of anomalies. When video is being monitored, the system only needs to compare current footage against these pre-prepared models and baselines, significantly reducing real-time processing requirements and detection time while maintaining comprehensive analysis capability.
3Area of stationary object
If the number of monitored cameras is increased, then monitoring coverage is improved, but the number of videos to confirm increases making it more difficult to monitor all videos
Solution Approach 1:
The system implements self-service by automatically prioritizing and selecting which camera feeds require observer attention based on real-time abnormality detection. The monitoring system serves itself by autonomously filtering, ranking, and presenting only the most relevant video feeds to the observer, eliminating the need for manual review of all camera feeds and enabling effective monitoring of large numbers of cameras without increasing operational difficulty.
4Stability of the object's composition
If observer switches videos every predetermined time interval, then systematic monitoring is maintained, but important camera videos may be missed
Solution Approach 1:
The system transitions from static, fixed-interval video switching to dynamic, adaptive video selection based on real-time abnormality detection. The system continuously adjusts which videos are prioritized for display based on detected abnormality levels, allowing the monitoring focus to dynamically shift toward areas requiring attention while maintaining systematic coverage of all cameras through continuous background analysis of all feeds.
Data Source
AI summary
According to the present invention, switching of the monitoring images matching the intention of the observer can be automatically performed for images from a plurality of image capturing apparatus, and the load about the job of the observer can be reduced. The image monitoring apparatus includes an estimating unit configured to estimate attention degrees of a user for a plurality of images acquired from the plurality of image capturing apparatuses, a designating unit configured to designate one of the acquired images as an image to be displayed in accordance with an instruction from the user, a learning unit configured to cause the estimating unit to learn so as to increase an attention degree of the designated image, and a selecting unit configured to select one of the plurality of images based on an attention degree of each estimated image.