Finish-Line Camera Judging With AI Race Number Identification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for judging competitor positions and identities at a finish line in sports competitions, such as road cycling, are time-consuming and prone to errors due to manual operations and challenges in identifying race numbers, especially in dense groups, and systems like transponders can be inaccurate or fail.

Innovation Solution

A system and method using calibrated photo-finish and video cameras to automatically capture and process images, incorporating artificial intelligence and OCR, to quickly and accurately determine competitor positions and identities by enhancing images with a vertical line corresponding to the judging line.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If manual judgment methods are used, then equipment complexity is reduced, but judgment time and error rate increase

Engineering Contradiction:
Improvejudgment system complexityVSAvoidjudgment time
Core Design Contradiction:
Device complexityVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical judgment operations with an automated image processing system using cameras and computer vision algorithms. The system automatically detects competitors' positions, identifies race numbers, and determines finish order without human intervention, thereby eliminating the time loss associated with manual judgment while managing system complexity through automated processing.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Device complexity

If manual judgment methods are used, then system complexity is reduced, but measurement precision deteriorates

Engineering Contradiction:
Improvejudgment system complexityVSAvoidposition measurement accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent substitutes manual position marking with automated image processing that precisely measures competitor positions relative to the finish line. The system uses pixel-based coordinate detection and image enhancement algorithms to achieve high measurement precision for competitor positioning, race number identification, and finish order determination without the errors inherent in manual operations.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If transponder systems are used, then identification speed is improved, but reliability deteriorates due to equipment changes and inaccurate data

Engineering Contradiction:
Improveidentification speedVSAvoididentification accuracy
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent uses optical copying through photo-finish cameras to capture images of competitors and their race numbers at the finish line. Instead of relying on transponders that can be removed or malfunction, the system creates visual copies of the competitors' identification elements (race numbers, uniforms) directly from the finish line images, ensuring reliable identification that is not affected by equipment changes during the event.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent introduces photo-finish images as an intermediary between the competitors and the judgment system. The images serve as a permanent record that can be analyzed multiple times by automated algorithms to extract identification data, providing a reliable intermediary layer that is not susceptible to the failures of direct transponder reading, especially in cases of equipment changes or poor signal reception.

Inventive Principle:
Principle #24Intermediary (Mediator)

4Measurement precision

If photo-finish images are enhanced with vertical lines, then measurement precision is improved, but device complexity increases

Engineering Contradiction:
Improveposition measurement accuracyVSAvoidimage processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the image processing task into distinct functional steps: capturing raw photo-finish images, enhancing image quality, superimposing vertical reference lines at predetermined positions, detecting competitor positions relative to these lines, and determining finish order. This segmentation allows each step to be optimized independently, improving measurement precision through structured processing while managing overall system complexity through modular design.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4679389A1System and method for judging at least one competitor passing over a judging line
Publication Date: 2026.01.14 SWISS TIMING LTD
  • EP4679389A1 patent drawingFigure 1
  • EP4679389A1 patent drawingFigure 2
  • EP4679389A1 patent drawingFigure 3

AI summary

The invention relates to a system and a method for judging at least one competitor as he crosses a judging line, such as a finishing line (50), of given width, the system comprising at least one camera (10, 20) for taking photo-finish images, disposed aligned with the judging line, the system comprising at least one video camera (30) disposed above the judging line.