Improving Map Maintenance Efficiency with Building Change Detection AI

Background

Maintaining nationwide map data requires efficient and comprehensive detection of changes across vast areas.

Geotechnologies, Inc. previously updated its map data by acquiring large volumes of satellite imagery and combining AI analysis with manual visual inspection. However, challenges remained in sourcing and managing the required data efficiently.

Challenges Before Implementation

In map data maintenance, target areas were traditionally selected based on past update patterns and operational experience. Aerial imagery and optical satellite data were then acquired, and map updates were performed based on visual interpretation of these images.

However, the acquired data often included areas with little or no observable change, resulting in unnecessary costs. In addition, the selection of target areas relied heavily on individual expertise, leading to inconsistencies and inefficiencies in the process.

As a result, a key challenge was to identify areas of change in advance and optimize data acquisition accordingly.

Implementation Results

The Building Change Extraction Solution, introduced in FY2024, is a service powered by our Building Change Detection AI. It identifies building changes—such as new construction and demolitions—at a nationwide scale, and quantitatively visualizes areas with high and low levels of change. This enables more effective prioritization of map updates.

By implementing this solution, Geotechnologies, Inc. reduced overall map maintenance costs, including data acquisition, by 13.2%. In addition, the rate of map updates within acquired data improved by approximately 1.8×, resulting in significant gains in both operational efficiency and data quality (Figure 1).

Figure 1. Impact of the Building Change Extraction Solution
Source: Geotechnologies, Inc.

1) 1.8× Increase in Update Efficiency

The proportion of buildings updated increased from 5.7% to 10.1%, driven by targeted data acquisition in high-change areas.

2) 13.2% Reduction in Maintenance Costs

Maintenance costs, including data procurement, were reduced by 13.2% through targeted data acquisition in high-change areas.

3) Improved Planning Accuracy with Objective Metrics

The use of change intensity as an objective metric enabled more accurate area selection and nationwide prioritization of map updates.

Summary

By implementing the Building Change Extraction Solution powered by our Building Change Detection AI, Space Shift helped Geotechnologies, Inc. address key challenges in cost efficiency, update rates, and planning accuracy. This also contributed to the establishment of a standardized map maintenance workflow at a nationwide scale.

Beyond map maintenance, this approach has broad applicability across urban planning, disaster management, and infrastructure monitoring, demonstrating the potential of satellite data as a scalable and practical solution.

If you are interested in other use cases or applications of Building Change Detection AI, please feel free to contact us.

Learn more about Building Change Detection AI

Inquiries and Document Request

SpaceShift Inc.
Please feel free to contact us with any questions regarding the utilization of satellite data for agriculture and green growth, requests for detailed explanations, or consultations regarding implementation.
[Email] sales@spcsft.com