Vehicle Image Detection by State-Specific Model Selection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The accuracy of detecting vehicles in manufacturing environments is compromised due to variations in appearance across different manufacturing steps, leading to decreased detection precision.

Innovation Solution

A detecting apparatus utilizing machine learning models to classify vehicle states and select appropriate detection models based on captured image analysis, incorporating distortion correction and orientation processing to enhance detection accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a single detection model is used for all vehicle appearances, then device complexity is reduced, but detection precision deteriorates due to appearance variations across manufacturing steps

Engineering Contradiction:
Improvedetection precisionVSAvoiddetection model complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the detection system into multiple specialized detection models, each trained for specific vehicle appearances at different manufacturing steps. This segmentation allows each model to focus on particular characteristics, improving detection precision while managing complexity through modular architecture

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system dynamically selects the appropriate detection model based on the current vehicle appearance and manufacturing step. This dynamic adaptation enables the system to optimize detection precision for each specific scenario without being constrained by a single static model

Inventive Principle:
Principle #15Dynamics

2Measurement precision

If multiple detection models are prepared for different vehicle states, then detection precision is improved, but processing time increases due to model selection and switching

Engineering Contradiction:
Improvevehicle detection accuracyVSAvoiddetection processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary classification of vehicle appearance and manufacturing step before detection. This preliminary action enables rapid model selection, reducing the time penalty associated with having multiple detection models by pre-organizing the selection process

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediate classification mechanism that bridges the gap between diverse vehicle appearances and specific detection models. This intermediary layer efficiently routes input images to the appropriate model, minimizing selection time and maintaining high detection accuracy

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4481692A1Detecting apparatus, position calculation system, and detecting method
Publication Date: 2024.12.25 TOYOTA JIDOSHA KK
  • EP4481692A1 patent drawingFigure 1
  • EP4481692A1 patent drawingFigure 2
  • EP4481692A1 patent drawingFigure 3~4

AI summary

In a detecting apparatus (5; 5a; 5c; 5d) configured to detect a vehicle (10) in a captured image, the vehicle (10) is configured to move in a factory in which manufacturing steps are performed to manufacture and ship the vehicle (10), the vehicle (10) is classified into states in accordance with an appearance of the vehicle (10), varying among the manufacturing steps. The detecting apparatus (5; 5a; 5c; 5d) includes a first processor (52; 52c; 52d) configured to acquire the captured image, acquire state information indicating one of the states of the vehicle (10) in the captured image, from among first detection models that are machine learning models each prepared one by one for the states, acquire the first detection model selected in accordance with the one of the states, identified by the acquired state information, and detect the vehicle (10) in the captured image by inputting the captured image to the acquired first detection model and identifying a target region representing the vehicle (10) in the captured image.