NASA · NASA-900 · record R0497

GLOBE Program - Image processing

National Aeronautics and Space Administration · GSFC: Goddard Space Flight Center

Overview

Development stage: DeployedSource: c) Deployed – The use case is being actively authorized or utilized to support the functions or mission of an agency.
High-impact designation: Not high-impactSource: c) Not high-impact
Impact justification: Not reported
Start date: 2022-03-29 (day precision)Source: 03/29/2022 00:00:00
Withheld from public reporting: NoSource: a) No

Mission

Topic area: Science
Operational functions: Image and video analysis Derived by keyword rules; see methodology

Problem the AI is intended to solve

Processing photo images submitted to the GLOBE Program through the GLOBE Observer app. These photos relate to submissions of Clouds, Mosquito Habitat Mapper, Trees, and Land Cover observations. GLOBE data is processed through Amazon Rekognition SAAS (software as a service) tool to automate the image approval process (Amazon 2024).

Expected benefits

The AI photo review system has dramatically reduced staff time needed to screen photos and therefore enabled NASA to process these crowdsourced data at a much faster speed. This leads to significantly more surface-based Earth system observations to enable science and societally relevant applications that would not be possible otherwise.

System outputs

Object detection information, image approval/rejection decisions, automated blurring of detected faces and text for privacy protection, and flagging of images requiring human review.

Technology

AI classification: Computer visionSource: Computer Vision: AI that processes and interprets visual data (e.g., images and videos).
System name(s): AWS
Custom-developed code: Yes
Public source code: Not reported

Sourcing

How it was built: Contract and in-houseSource: c) Developed with both contracting and in-house resources
Vendor (as reported): Mitchell Vantage Systems (MVS), subcontracted to Astrion
Vendors (standardized): None on the standard list

A vendor is the supplier named by the agency. It does not identify the underlying model or AI technology.

Data and privacy

Involves PII: No
Privacy Impact Assessment: Not reported
Authorization to Operate: Yes
Demographic features: Not reported
Training and evaluation data: Mostly uses pre-trained Amazon Rekognition models. System processes citizen science photos for validation but does not train custom models. However, a custom process was developed to evaluate the Rekognition results for decision making within the GLOBE context.
Federal Data Catalog: Not reported

Governance

0 of 8 minimum-practice questions answered. A blank answer means the agency reported nothing; it does not mean the practice is absent.

Pre-deployment testingNo answer reported

AI impact assessmentNo answer reported

Independent reviewNo answer reported

Ongoing monitoringNo answer reported

Operator trainingNo answer reported

Fail-safeNo answer reported

Appeal processNo answer reported

User and public consultationNo answer reported

Potential impacts and how they were identified

Not reported