Title :
Cascade classifier: design and application to digit recognition
Author :
Duong, Jean ; Emptoz, Hubert
Author_Institution :
Lab. d´´InfoRmatique en Images et Systemes d´´information, Inst. Nat. des Sci. Appliquees de Lyon, Villeurbanne, France
fDate :
29 Aug.-1 Sept. 2005
Abstract :
The focus of this paper is on machine learning. More specifically, a classifier combination is proposed. We name it "cascade classifier". Conceptually, this system is based on confusion analysis and local retraining mechanism. We describe its architecture. Experiments with handwritten digit images of the MNIST ("modified NIST") database are presented.
Keywords :
handwritten character recognition; learning (artificial intelligence); pattern classification; visual databases; cascade classifier; classifier combination; confusion analysis; digit recognition; handwritten digit image; local retraining mechanism; machine learning; modified NIST database; Computer applications; Computer architecture; Covariance matrix; Focusing; Fuzzy systems; Image databases; Machine learning; Neural networks; Pattern recognition; Principal component analysis;
Conference_Titel :
Document Analysis and Recognition, 2005. Proceedings. Eighth International Conference on
Print_ISBN :
0-7695-2420-6
DOI :
10.1109/ICDAR.2005.69
