Semi-autonomous on-line studying method based on random fern classifier

Abstract

The invention provides a semi-autonomous on-line studying method based on a random fern classifier. The semi-autonomous on-line studying method based on the random fern classifier comprises the following steps that (1) a sample set for initial training of the classifier is prepared; (2) initial training of the random fern classifier is carried out; (3) on-line study samples are obtained; (4) on-line training of the random fern classifier is carried out. By means of the semi-autonomous on-line studying method based on the random fern classifier, the performance of a target detection system can be improved gradually, and the requirement for the detection accuracy is finally met. In addition, correctness of category marking of the on-line study samples is guaranteed through the method of obtaining the on-line study sample through manual judgment. Compared with a traditional classifier training method, the semi-autonomous on-line studying method based on the random fern classifier does not need ahead preparation of a large number of positive and negative training samples, lightens the workload of manual marking, and can verify the classification performance of the random fern classifier.

Description

technical field [0001] The invention belongs to a pattern recognition method, and in particular relates to a method for training a classifier through an online learning algorithm to improve the performance of the classifier. Background technique [0002] Online learning belongs to the research category of incremental learning. In this type of method, the classifier only learns each sample once instead of repeated learning. In this way, a large amount of storage space is not required to store training samples during the operation of the online learning algorithm. Every time the classifier obtains a sample, it conducts online learning. Through online learning, the classifier can still update and improve itself according to the new sample during use, and further improve the classification effect. [0003] Early online learning algorithms include Winnow algorithm, unified linear prediction algorithm, etc. In 2001, scholar Oza combined these algorithms with boosting algorithm and...

AI Technical Summary

Problems solved by technology

However, online Adaboost uses feature selection operators instead of general weak classifiers to synthesize strong classifiers. The number of feature selection operators and the number of weak classifiers corresponding to feature selection operators are fixed, and the corresponding online learning classifier structure is relatively rigid.

When it is found that its classification ability cannot meet the requirements of detection performance, even continuous online learning cannot improve the detection accuracy

Images