Quantifying the unimportance of prior probabilities in a computer vision problem

We present an empirical investigation of the importance of accurate assessment of prior probabilities in a typical visual classification problem, handwritten ZIP Code recognition. We found that little accuracy was gained by accurate assessment of the prior distribution over the individual digits.

Saved in:
Bibliographic Details
Published in:Pattern recognition letters 1990, Vol.11 (6), p.379-383
Main Authors: Sher, David B., Hull, Jonathan J.
Format: Article
Language:English
Subjects:
Applied sciences
Artificial intelligence
Bayesian inference
character recognition
Computer science
control theory
systems
computer vision
decision theory
Exact sciences and technology
image analysis
pattern classification
Pattern recognition. Digital image processing. Computational geometry
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present an empirical investigation of the importance of accurate assessment of prior probabilities in a typical visual classification problem, handwritten ZIP Code recognition. We found that little accuracy was gained by accurate assessment of the prior distribution over the individual digits.
ISSN:0167-8655
1872-7344
DOI:10.1016/0167-8655(90)90107-D