Scoreboard Region Detection for Efficient Sport Data Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for extracting sport-related information from media streams, such as video feeds of sporting events, are inefficient, computationally expensive, and prone to errors due to the recognition of unrelated text elements, requiring additional processing to isolate relevant information.
Innovation Solution
A computing system automatically analyzes video frames to identify a scoreboard region based on static image elements, detects sub-regions containing text, and applies optical character recognition (OCR) to generate sport-related data, focusing on relevant frame regions by leveraging the unchanging nature of scoreboard elements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If OCR is applied to the entire video frame to extract sport data, then comprehensive text recognition is achieved, but computational overhead increases and recognition accuracy decreases due to unrelated text elements
Solution Approach 1:
The patent divides the video frame into multiple candidate regions based on spatial coordinates, then identifies the scoreboard region by analyzing static elements across multiple frames. This segmentation approach focuses OCR processing only on the relevant scoreboard area rather than the entire frame, improving both accuracy and efficiency
Solution Approach 2:
The patent applies different processing strategies to different regions of the video frame. By identifying the scoreboard region as a specific local area with distinct characteristics (static elements, specific coordinate ranges), the system applies OCR only to this region, thereby improving recognition accuracy while reducing computational overhead
2Loss of information
If the entire video frame is analyzed to identify scoreboard text, then all text elements are captured, but processing time increases and relevant information is lost among unrelated text
Solution Approach 1:
The patent performs preliminary analysis by identifying candidate regions and determining the scoreboard region before applying OCR. This preliminary action involves analyzing static elements across multiple frames to establish the scoreboard's spatial coordinates, thereby preparing the system to quickly extract only relevant sport data without analyzing the entire frame
Solution Approach 2:
The patent extracts the scoreboard region from the entire video frame by identifying its specific spatial coordinates based on static elements. This extraction isolates the relevant sport data containing area from unrelated text and visual elements, enabling focused processing that reduces time loss while maintaining data completeness
3Reliability
If manual methods are used to record sport event information, then accurate data collection is achieved, but time consumption and cost increase significantly
Solution Approach 1:
The patent implements an automated system that performs sport data extraction independently without human intervention. The system automatically identifies the scoreboard region, extracts text using OCR, and processes sport data, thereby eliminating the time consumption and costs associated with manual recording while maintaining high accuracy through algorithmic precision
4Productivity
If OCR is applied without region identification to extract sport data, then all text in the frame is recognized, but computational resources are wasted on unrelated text elements
Solution Approach 1:
The patent segments the video frame into candidate regions and identifies the scoreboard region as the specific area containing sport data. This segmentation enables the system to apply OCR only to the relevant region, thereby improving extraction speed while reducing computational resource consumption by excluding unrelated text elements from processing
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
A computing system engages in digital image processing of received video frames to generate sport data that indicates a score and/or a time associated with a sport event. The digital image processing includes: (i) identifying a first frame region of the video frames based on the first frame region depicting a scoreboard; (ii) executing a first procedure that analyzes the identified first frame region to detect, within the identified first frame region, second frame region(s) based on the second frame region(s) depicting text of the scoreboard: (iii) m response to detecting the second frame region(s), executing a second procedure to recognize the text in at least one of the second frame region(s); and (iv) based at least on the recognizing of the text, generating the sport data. In response to completing the digital image processing, the computing system then carries out an action based on the generated sport data.