Actions for Unsupervised texture image segmentation by improved neural network ART2

Email RIS file
Bookmark
Report an Issue

Unsupervised texture image segmentation by improved neural network ART2

Author
Labini, G. Sylos
Published
Mar 1, 1994.
Physical Description
1 electronic document
Additional Creators
Desario, Marco, Mugnuolo, R., and Wang, Zhiling
Online Version

Availability

I Want It
I Want It

Finding items...

Restrictions on Access
Unclassified, Unlimited, Publicly available.
Free-to-read Unrestricted online access
Summary
We here propose a segmentation algorithm of texture image for a computer vision system on a space robot. An improved adaptive resonance theory (ART2) for analog input patterns is adapted to classify the image based on a set of texture image features extracted by a fast spatial gray level dependence method (SGLDM). The nonlinear thresholding functions in input layer of the neural network have been constructed by two parts: firstly, to reduce the effects of image noises on the features, a set of sigmoid functions is chosen depending on the types of the feature; secondly, to enhance the contrast of the features, we adopt fuzzy mapping functions. The cluster number in output layer can be increased by an autogrowing mechanism constantly when a new pattern happens. Experimental results and original or segmented pictures are shown, including the comparison between this approach and K-means algorithm. The system written in C language is performed on a SUN-4/330 sparc-station with an image board IT-150 and a CCD camera.
Other Subject(s)
Collection
NASA Technical Reports Server (NTRS) Collection.
Note
Document ID: 19940026044.
Accession ID: 94N30549.
AIAA PAPER 94-1199-CP.
NASA. Johnson Space Center, Conference on Intelligent Robotics in Field, Factory, Service, and Space (CIRFFSS 1994), Volume 1; p 175-180.
Terms of Use and Reproduction
No Copyright.
View MARC record | catkey: 15659088