Broadcast Content Detection via Self-Similarity Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radio monitoring systems fail to detect and track unknown broadcast content items that have not been registered in their databases, as they rely solely on matching incoming signals with pre-existing pattern vectors, resulting in these unknown items being ignored.
Innovation Solution
The system employs the self-similarity principle to detect repeated programs, even if they are not registered, by extracting pattern vectors from repeated instances and using them to identify and register unregistered content, with an Audio-Selector choosing the best-quality repetition for database entry, and operates in modes that focus on undetected or all time periods for comprehensive content harvesting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system relies solely on matching incoming signals with pre-existing pattern vectors in the database, then the detection process is simple and fast, but unknown or unregistered content items are ignored and cannot be detected
Solution Approach 1:
The system performs preliminary detection of potential content items and stores them as candidate patterns before formal registration. This preliminary action allows the system to capture unknown content that hasn't been officially registered yet, while maintaining the structured approach of pattern matching for known content.
Solution Approach 2:
The patent introduces an intermediary mechanism between known content and unknown content detection. The system uses a candidate pattern buffer that acts as an intermediary storage layer, allowing transitions from detecting only registered content to detecting and tracking unknown content, and finally to formal registration.
2Adaptability or versatility
If the system tracks and submits every potential unknown content item for human identification, then detection of unregistered content improves, but system complexity and processing time increase
Solution Approach 1:
The system applies partial action by not submitting every single detected pattern for human identification. Instead, it uses filtering criteria to select only those candidate patterns that meet certain thresholds for uniqueness and quality, reducing the burden on human operators while still capturing important unknown content.
Solution Approach 2:
The system implements feedback mechanisms where detection results are analyzed and used to adjust detection parameters. The feedback loop allows the system to learn from previous detections and refinements, improving accuracy over time without requiring proportional increases in system complexity.
3Measurement precision
If the system processes all broadcast signals to detect unknown content, then content detection accuracy improves, but processing time and computational resources increase
Solution Approach 1:
The detection process is segmented into distinct phases: initial pattern extraction, candidate identification, filtering, and submission for identification. This segmentation allows the system to process signals efficiently by focusing computational resources on promising candidates rather than uniformly processing all possible content items.
Solution Approach 2:
The system performs preliminary filtering and analysis of broadcast signals to identify potential unknown content before committing to full processing. This preliminary action includes quick checks for pattern uniqueness and quality metrics, eliminating obvious false positives early in the process.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system and method of detecting unidentified broadcast electronic media content using a self-similarity technique is presented. The process and system catalogues repeated instances of content that has not been positively identified, but are sufficiently similar as to infer repetitive broadcasts. These catalogued instances may be further processed on the basis of different broadcast channels, sources, geographic locations of broadcasts or format to further assist the identification thereof.