Predictive Multi-Camera Surveillance Capture for Fast Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video surveillance systems often miss detecting fast-moving objects, objects outside the current field of view, or early portions of events due to delayed or inefficient video capture modifications, leading to incomplete data capture and increased storage costs.
Innovation Solution
A system that uses video data from a first camera to predictively modify video capture operations of another camera with non-overlapping fields of view, including adjusting capture rates, field of view direction, and storage modes based on detected events, before the object of interest enters the camera's view.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If video cameras selectively capture high quality video only when objects or motion are detected, then storage costs are reduced, but detection reliability deteriorates due to delayed object detection and missed fast moving objects
Solution Approach 1:
The system performs preliminary actions by having cameras continuously capture video at lower quality settings while maintaining readiness to switch to high quality capture. When a camera detects an object or motion, it proactively notifies neighboring cameras before those cameras need to capture the same event, allowing them to prepare and switch to high quality mode in advance, thus preventing missed detections while avoiding unnecessary continuous high quality recording
Solution Approach 2:
The system implements feedback mechanisms where detection results from one camera are communicated to neighboring cameras. When Camera A detects an object, it sends notification to Camera B (and other neighboring cameras), which then adjust their capture parameters based on this feedback information, improving overall system detection reliability without requiring all cameras to operate at full capacity simultaneously
2Loss of energy
If video cameras operate at lower capture rates to reduce data usage, then storage costs are reduced, but detection precision deteriorates by missing fast moving objects and early portions of events
Solution Approach 1:
Neighboring cameras receive advance notification of detected objects and proactively switch to high capture rate mode before the objects enter their fields of view. This preliminary action ensures that when fast-moving objects or critical event portions enter the camera view, the camera is already operating at sufficient capture rate to accurately detect and record them, preventing missed detections while maintaining lower average data usage
3Reliability
If a single camera continuously monitors all areas, then detection coverage is improved, but device complexity and storage requirements increase
Solution Approach 1:
The surveillance system is segmented into multiple cameras, each responsible for monitoring specific areas or directions. Rather than requiring one camera to monitor all areas, multiple cameras work in coordination, with each camera capturing events in its designated zone. This segmentation reduces individual camera complexity and storage requirements while maintaining or improving overall detection coverage through collaborative operation
Solution Approach 2:
Multiple cameras are merged into a coordinated system where detection information from one camera benefits neighboring cameras. The system combines the capabilities of individual cameras through inter-camera communication, allowing them to share detection results and adjust their operations accordingly, achieving comprehensive coverage equivalent to or better than a single omnidirectional camera while reducing total system complexity and storage requirements
Data Source
AI summary
Systems, video cameras, and methods for predictive adjustment of multi-camera surveillance video data capture are described. A plurality of networked video camera is deployed with known spatial relationships and non-overlapping fields of view among cameras. When a video event is detected from video data for one of the video cameras, a video capture update message is selectively sent to other video cameras to modify their video capture operations. For example, an object detected by one camera may trigger adjustment of video capture operating parameters for cameras in a direction of travel of the object before it has entered the field of view of those other cameras.