Automated Inspection of Parking Space Recognition Data

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The increasing volume of training and validation data for deep learning models in parking space recognition systems necessitates manual inspection, which is time-consuming and costly.

Innovation Solution

A computing device and method for automatically inspecting parking space recognition data using a data inspection device that determines position and type accuracy, and analyzes consistency, thereby reducing the need for manual intervention and improving data quality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual inspection of training data is performed, then data quality can be ensured, but inspection time and cost increase significantly

Engineering Contradiction:

Improvedata qualityVSAvoidinspection time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical inspection with an automated computer-based inspection system that uses algorithms to analyze training data quality, thereby eliminating the time-consuming manual process while maintaining inspection effectiveness

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The inspection system performs self-service by automatically detecting data quality issues without human intervention, using automated algorithms to identify problematic training data samples and generate inspection reports

Inventive Principle:

Principle #25Self-service

2Reliability

If manual inspection of training data is performed, then data quality can be ensured, but inspection cost increases

Engineering Contradiction:

Improvedata qualityVSAvoidinspection cost

Core Design Contradiction:

ReliabilityVSEase of manufacture

Solution Approach 1:

The patent replaces expensive manual inspection labor with an automated computer-based system, significantly reducing inspection costs while maintaining the ability to ensure data quality through algorithmic analysis

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The inspection system uses computationally efficient algorithms that can be executed quickly and cheaply on standard hardware, replacing the need for expensive human expert inspection resources

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

3Reliability

If deep learning model uses more training data, then model performance improves, but inspection complexity increases

Engineering Contradiction:

Improvemodel performanceVSAvoidinspection complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the inspection process into modular components that can handle different aspects of data quality assessment independently, making the overall inspection system more manageable and scalable for large datasets

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The inspection system is designed with universal algorithms that can handle various types of training data and quality issues through a single unified platform, reducing inspection complexity despite increasing data volume

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20250124722A1Computing apparatus and method for inspecting learning data

Publication Date: 2025.04.17 HYUNDAI MOTOR CO LTD

AI summary

A computing device includes a parking space recognition device that recognizes a parking space using at least one parking space recognition model for recognizing the parking space based on training data for the parking space and a data inspection device that inspects information about a parking line and a parking slot based on parking space recognition data recognized by the parking space recognition device.