Queue Analysis by Image-Based Line Generation Without Shape Input

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image processing systems require advance knowledge of the queue shape to detect and analyze queues, necessitating user input for shape recognition.

Innovation Solution

A queue analysis apparatus that estimates the position and orientation of objects in an image and generates a queue line based on these estimates, eliminating the need for prior knowledge of queue shape.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If advance knowledge of queue shape is required for detection, then detection accuracy can be improved, but user labor and operation complexity increase

Engineering Contradiction:

Improvequeue detection accuracyVSAvoiduser labor for shape recognition

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The system automatically determines queue shape and orientation without requiring user input. The estimation unit independently analyzes image data to extract queue characteristics, making the system self-sufficient and eliminating the need for manual shape recognition by users.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system performs preliminary estimation of queue shape and orientation parameters before actual queue detection. By pre-calculating these geometric characteristics from image data, the system prepares the necessary information in advance, enabling accurate detection without requiring user-provided shape knowledge.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If user provides advance knowledge of queue shape, then detection reliability improves, but device complexity and operation time increase

Engineering Contradiction:

Improvequeue detection reliabilityVSAvoidtime for shape recognition

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical shape recognition with automated computational estimation. The estimation unit uses image processing algorithms to automatically determine queue geometric characteristics, substituting human cognitive processing with machine-based automated analysis that is both reliable and time-efficient.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Ease of operation

If automated queue detection is implemented, then ease of operation improves, but measurement precision may deteriorate without shape knowledge

Engineering Contradiction:

Improveease of queue detectionVSAvoidqueue state recognition accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The system dynamically adapts its detection parameters based on the actual queue configuration detected in each image. Rather than relying on pre-defined static shape models, the estimation unit continuously adjusts its analysis to match the real-time geometric characteristics of queues in the image data, maintaining high precision across diverse queue scenarios.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS12406465B2Queue analysis apparatus, control method, and non-transitory computer-readable medium

Publication Date: 2025.09.02 NEC CORP

AI summary

A queue analysis apparatus (2000) estimates a position and an orientation of each object (20) included in a target image (10). The target image (10) is generated by a camera (50) that captures the object (20). The queue analysis apparatus (2000) generates a queue line (40) that expresses, by a linear shape, a queue included in a queue region (30) being a region representing a queue in the target image (10), based on a position and an orientation being estimated for each object (20) included in the queue region (30).