DocumentCode :
3410722
Title :
Combining image classification and image compression using vector quantization
Author :
Oehler, Karen L. ; Gray, Robert M.
Author_Institution :
Dept. of Electr. Eng., Stanford Univ., CA, USA
fYear :
1993
fDate :
1993
Firstpage :
2
Lastpage :
11
Abstract :
The goal is to produce codes where the compressed image incorporates classification information without further signal processing. This technique can provide direct low level classification or an efficient front end to more sophisticated full-frame recognition algorithms. Vector quantization is a natural choice because two of its design components, clustering and tree-structured classification methods, have obvious applications to the pure classification problem as well as to the compression problem. The authors explicitly incorporate a Bayes risk component into the distortion measure used for code design in order to permit a tradeoff of mean squared error with classification error. This method is used to analyze simulated data, identify tumors in computerized tomography lung images, and identify man-made regions in aerial images
Keywords :
Bayes methods; computerised tomography; image coding; medical image processing; vector quantisation; Bayes risk component; aerial images; classification error; clustering; code design; computerized tomography; full-frame recognition algorithms; image classification; image compression; mean squared error; tree-structured classification; vector quantization; Classification tree analysis; Clustering algorithms; Computer errors; Data analysis; Distortion measurement; Image analysis; Image classification; Image coding; Signal processing algorithms; Vector quantization;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Data Compression Conference, 1993. DCC '93.
Conference_Location :
Snowbird, UT
Print_ISBN :
0-8186-3392-1
Type :
conf
DOI :
10.1109/DCC.1993.253150
Filename :
253150
Link To Document :
بازگشت