Accession Number:

ADA106814

Title:

Texture Classification with Change Point Statistics.

Descriptive Note:

Technical rept.,

Corporate Author:

MARYLAND UNIV COLLEGE PARK COMPUTER VISION LAB

Personal Author(s):

• Dunn,Stanley M.

Report Date:

1981-07-01

Pagination or Media Count:

19.0

Abstract:

Although the problem of texture classification has been thoroughly studied, it is still desirable to investigate whether or not we can model the random behavior of the imaging equipment and still achieve a reliable classification scheme. To this end, we introduce a statistical test previously used to determine whether or not a sequence of random variables has a change point.

Descriptors:

• *IMAGE PROCESSING
• *STATISTICAL ANALYSIS
• *PHOTOGRAPHIC TEXTURE
• STATISTICAL TESTS
• RANDOM VARIABLES
• GAUSSIAN NOISE
• NONPARAMETRIC STATISTICS
• RELIABILITY
• CLASSIFICATION
• PATTERN RECOGNITION
• SEQUENCES(MATHEMATICS)
• POINTS(MATHEMATICS)

Subject Categories:

• Statistics and Probability

Distribution Statement:

APPROVED FOR PUBLIC RELEASE