VARIANCE PREDICTION FOR POPULATION SIZE ESTIMATION

Ana Isabel Gomez, Marcos Cruz, Luis Manuel Cruz-Orive

Abstract

Design unbiased estimation of population size by stereological methods is an efficient alternative to automatic computer vision methods, which are generally biased. Moreover, stereological methods offer the possibility of predicting the error variance from a single sample. Here we explore the statistical performance of two alternative variance estimators on a dataset of 26 labelled crowd pictures. The empirical mean square errors of the variance predictors are compared by means of Monte Carlo resampling.


Keywords
Cavalieri error variance predictor; geometric sampling; Monte Carlo resampling; particle counting; population size; Split error variance predictor; systematic quadrats

Full Text:

PDF


DOI: 10.5566/ias.1991

Copyright (c) 2019 Image Analysis & Stereology

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Image Analysis & Stereology
EISSN 1854-5165 (Electronic version)
ISSN 1580-3139 (Printed version)