Sensor-Triggered Image Capture for License Plate Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing law enforcement systems struggle to efficiently capture and process still images of regions-of-interest, such as vehicle license plates, during video streaming, especially under challenging conditions, which hinders efficient law enforcement operations.

Innovation Solution

A computer-implemented method and network server system that captures still images during video streaming using sensors and algorithms to identify regions-of-interest, applying optical character recognition (OCR) to extract patterns like license plate characters, with event detection models to trigger image capture and enhance resolution as needed.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If still images are captured continuously during video streaming to ensure pattern identification, then identification accuracy is improved, but data transmission load and processing time increase

Engineering Contradiction:
Improvepattern identification accuracyVSAvoidimage processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary actions by continuously capturing still images during video streaming before pattern identification is needed. This ensures that high-quality images are already available when pattern identification is triggered, eliminating the need for delayed capture and reducing overall processing time.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system uses self-service by automatically triggering pattern identification on captured still images without requiring manual intervention. The event detection model autonomously determines when pattern identification should occur, and the system automatically processes the images, reducing both human workload and processing delays.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If high resolution still images are captured for pattern identification, then identification accuracy is improved, but bandwidth consumption and storage requirements increase

Engineering Contradiction:
Improvelicense plate character recognition accuracyVSAvoiddata transmission volume
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The system extracts only the essential information from high-resolution still images for pattern identification, such as license plate regions and key visual features. By extracting and transmitting only the critical data elements rather than entire high-resolution images, the system maintains identification accuracy while significantly reducing data transmission volume.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system applies local quality by capturing high-resolution images only in specific regions of interest (such as license plate areas) rather than uniformly high resolution across the entire image. This allows the system to maintain high identification accuracy for critical elements while reducing overall data transmission requirements.

Inventive Principle:
Principle #3Local quality

3Measurement precision

If manual review of captured images is required for pattern identification, then accuracy is improved, but operational efficiency decreases

Engineering Contradiction:
Improvepattern identification accuracyVSAvoidlaw enforcement operational efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system implements self-service through automated pattern identification using event detection models and OCR technology. The system autonomously captures still images, identifies patterns such as license plate characters, and generates identification results without requiring manual review, thereby maintaining high accuracy while dramatically improving operational efficiency.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system replaces the mechanical process of manual image review with automated computational methods including event detection models and optical character recognition (OCR). This substitution eliminates human intervention in the pattern identification process, maintaining or improving accuracy while significantly increasing productivity and operational efficiency.

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

Data Source

PatentEP4356347B1Techniques for capturing enhanced images for pattern identifications
Publication Date: 2025.11.05 GETAC HLDG CORP
  • EP4356347B1 patent drawingFigure 1
  • EP4356347B1 patent drawingFigure 2
  • EP4356347B1 patent drawingFigure 3

AI summary

This disclosure describes techniques for capturing still images during video streaming to identify patterns in a region-of-interest on the captured still images. The video streaming may be performed by an imaging device that includes or is communicatively connected to one or more sensors (e.g., radar, light sensor, etc.) The one or more sensors may be configured to perform data measurements such as vehicle speed measurements, light intensity measurements, and/or the like. In one example, during the video streaming, the data measurement may be compared with a corresponding threshold. In this example, the imaging device may be triggered to capture still images of the surrounding area based on the comparison between the data measurement and the corresponding threshold. Thereafter, the still images may be processed to identify the region-of-interest on the still images.