Parking Lot Area Estimation from Vehicle Position Data

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for organizing map data of parking lots are inefficient, especially for large numbers of parking lots within a single block, as they require manual drawing and lack effective techniques for estimating parking lot areas.

Innovation Solution

An information processing device and method that acquires position information from movable bodies to estimate parking lot areas by analyzing their movements and parking positions, distinguishing between running and parking areas, and updating parking lot information based on density and time data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If manual drawing methods are used to organize map data of parking lots, then map data can be created, but the productivity is extremely low especially for huge number of parking lots

Engineering Contradiction:

Improveproductivity of organizing map dataVSAvoidtime required for manual map drawing

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical drawing operations with automated image processing and recognition systems. The system automatically detects parking lot boundaries from aerial images or map data through computer-based image processing, eliminating the need for workers to manually draw each parking lot area.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent uses existing aerial images and map data as source material to automatically generate parking lot boundary information. Instead of creating map data from scratch through manual drawing, the system copies and processes existing visual data to extract parking lot areas, significantly improving efficiency.

Inventive Principle:

Principle #26Copying

2Measurement precision

If pseudo parking lot areas enclosed by roads are used, then some parking lot coverage is achieved, but the measurement precision is insufficient when there are multiple parking lots in a single block

Engineering Contradiction:

Improveprecision of parking lot area identificationVSAvoidability to handle multiple parking lots in single block

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent applies segmentation by dividing the image processing into distinct stages: detecting candidate areas enclosed by roads, then further segmenting these areas to identify individual parking lot boundaries within each enclosed space. This allows multiple parking lots within a single block to be separately identified and measured accurately.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by using different processing strategies for different areas: first identifying road-enclosed regions globally, then applying specific parking lot detection algorithms locally within each region to distinguish individual parking lots based on their unique characteristics such as entrance positions and internal road patterns.

Inventive Principle:

Principle #3Local quality

Data Source

PatentEP3550538B1Information processing apparatus, information processing method, and program

Publication Date: 2025.01.01 PIONEER IP

Figure 1A~1B
Figure 2
Figure 3

AI summary

The parking lot area is estimated based on the information acquired by driving vehicles. The information processing device acquires position information including a parking position from a movable body, and estimates parking lot information related to a parking lot existing in an out-of-road area based on the position information of plural movable bodies entered the out-of-road area. By analyzing movements and/or parking positions of the movable bodies in the out-of-road area, information related to the parking lot can be estimated.