Back to search

South Dakota State University

Methods for Detecting Data Corruption in AVHRR Imagery

Abstract

dc:description.abstract

During the 1980’s researchers began to explore the use of relatively low resolution data sets to monitor large geographical areas. The primary interest was to improve mapping, resource analysis, and land management activities. One of the sources of this data is from the Advanced Very High Resolution Radiometer, AVHRR, carried aboard satellite platforms operated by the National Oceanic and Atmospheric Administration, NOAA. These sensors are capable of providing a maximum of 1.1 km resolution at the orbital subpoint. The corresponding size of the generated data set is about 100 million bytes for a five band multi-spectral observation covering 10 million square kilometers of Earth geography. One of the more challenging problems associated with the computer processing of these large data sets is how to detect data corruption that may occur due to electrical noise and hardware malfunctions in the complex electronics and communication systems used to transfer the data from satellite platforms to the ultimate end users. This paper examines several of the more common disruption situations and presents a technique to efficiently search for corrupted data. A single pass, two stage, data search procedure is presented which utilizes statistical methods to filter out the corrupted data. One stage of the process, based on global line statistics, is designed to detect large scale problems. If global testing is inconclusive, a more localized inspection is performed to isolate regions of statistically outlying data. Simple analysis methods are then used to determine if there is a spatial pattern present indicating a corrupted region. Promising results from the prototype code included with this paper have prompted the EROS Data Center to begin formal testing and implementation of the ideas presented here within the framework of their AVHRR Data Acquisition and Processing System, ADAPS.

Degree

thesis:*
Name thesis:degree_name
Master of Science (MS)
Level thesis:degree_level
Thesis - University Access Only
Discipline thesis:degree_discipline
Electrical Engineering

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Dewald, James D.
Contributors dc:contributor
  • Dennis Helder

Subjects

dc:subject × 2

Rights

dc:rights
Language dc:language
English

Identifiers

dc:identifier.*
Repository record dc:identifier
https://openprairie.sdstate.edu/etd2/85

Chain of custody

source
Harvested from
South Dakota State University
Base URL
openprairie.sdstate.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Dewald, James D.. Methods for Detecting Data Corruption in AVHRR Imagery. Thesis - University Access Only thesis, https://openprairie.sdstate.edu/etd2/85