Surveillance Video Stream Facial Image Superimposition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Surveillance systems face high risks of system failure due to human error in tracking targets across multiple cameras, leading to potential illegal activities, property theft, attacks, and loss of life, as current technologies are prone to errors in maintaining clear facial views of targets.
Innovation Solution
A computer program product that receives and processes multiple video data streams, identifies inferior facial views, extracts a clear facial image from one stream, and superimposes it onto the target's location in another stream, ensuring a consistent and clear view of the target across all cameras, including mobile devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple video data streams are monitored manually to maintain clear facial views of targets, then surveillance coverage is comprehensive, but human error increases leading to system failure
Solution Approach 1:
The system automatically performs facial view assessment and superimposition operations without human intervention. The processor autonomously evaluates facial views across multiple video streams, identifies inferior views, and superimposes clear facial images onto target objects, eliminating the need for manual monitoring and reducing human error.
Solution Approach 2:
The patent replaces manual mechanical monitoring with an automated image processing system. The processor uses algorithms to automatically detect facial views, assess their quality, and perform superimposition, substituting human operators with an automated computational system that maintains higher reliability.
2Measurement precision
If clear facial images are obtained from multiple camera angles, then identification accuracy improves, but system complexity increases
Solution Approach 1:
The system extracts only the essential facial image data from multiple video streams rather than processing all visual information. By focusing specifically on facial regions and their view quality, the system achieves high identification accuracy while reducing the complexity of processing entire video streams.
Solution Approach 2:
The system performs preliminary assessment of facial views across all video streams before final identification. By pre-evaluating which streams provide acceptable facial views and which require superimposition, the system simplifies the subsequent processing steps and maintains high identification accuracy efficiently.
3Speed
If real-time superimposition of facial images is performed, then surveillance response time improves, but processing load increases
Solution Approach 1:
The system performs superimposition only when necessary - specifically when a video stream provides an inferior facial view. Rather than processing and superimposing images from all streams continuously, the system selectively applies the technique only to streams that require enhancement, reducing overall processing load while maintaining fast response times.
Solution Approach 2:
The processor acts as an intermediary that intelligently selects which video streams require superimposition based on facial view assessment. This mediation prevents unnecessary processing of already-acceptable streams, balancing response speed with energy consumption by applying computational resources only where needed.
Data Source
AI summary
Systems and computer program products provide surveillance including a modified video data stream. The systems and products include computer readable program code, when read by a processor, that is configured for receiving at an image processor a first video data stream and a second video data stream, each of the first and second video data streams may include a target object having an assigned tracking position tag. The code further includes extracting a first facial image of the target object from the first video data stream, determining a target object location in the second video data stream based at least in part on the tracking position tag and generating a modified video data stream including the first facial image superimposed on or adjacent to the target object location in the second video data stream.