A Model Based Framework for Object Detection via Data Transformation

Peifang Guo

ISBN 10: 3659494577 / ISBN 13: 9783659494574

Published by LAP LAMBERT Academic Publishing Dez 2013, 2013

About this Item

This item is printed on demand - Print on Demand Titel. Neuware -The dynamic synthesis of nonlinear feature functions is a challenging problem in object detection. This book presents a combinatorial approach of genetic programming and the expectation maximization algorithm (GP-EM) to synthesize nonlinear feature functions automatically for the purpose of object detection. The EM algorithm investigates the use of Gaussian mixture which is able to model the behaviour of the training samples during an optimal GP search strategy. Based on the Gaussian probability assumption, the GP-EM method is capable of performing simultaneously dynamic feature synthesis and model-based generalization. The EM part of the approach leads to the application of the maximum likelihood (ML) operation which provides protection against inter-cluster data separation and thus exhibits improved convergence. The experimental results show that the approach improves the detection accuracy and efficiency of pattern object discovery, as compared to some state-of-the-art methods for object detection existing in the literature.VDM Verlag, Dudweiler Landstraße 99, 66123 Saarbrücken 64 pp. Englisch.

Seller Inventory # 9783659494574

Bibliographic Details

• Title: A Model Based Framework for Object Detection via Data Transformation
• Publisher: LAP LAMBERT Academic Publishing Dez 2013
• Publication Date: 2013
• Language: English
• ISBN 10: 3659494577
• ISBN 13: 9783659494574
• Binding: Taschenbuch
• Condition: Neu
• Dimensions: 220W x 150H x 5D millimeters 220 millimeters width by 150 millimeters height by 5 millimeters depth
• Weight: 113 grams

About this title

Synopsis

The dynamic synthesis of nonlinear feature functions is a challenging problem in object detection. This book presents a combinatorial approach of genetic programming and the expectation maximization algorithm (GP-EM) to synthesize nonlinear feature functions automatically for the purpose of object detection. The EM algorithm investigates the use of Gaussian mixture which is able to model the behaviour of the training samples during an optimal GP search strategy. Based on the Gaussian probability assumption, the GP-EM method is capable of performing simultaneously dynamic feature synthesis and model-based generalization. The EM part of the approach leads to the application of the maximum likelihood (ML) operation which provides protection against inter-cluster data separation and thus exhibits improved convergence. The experimental results show that the approach improves the detection accuracy and efficiency of pattern object discovery, as compared to some state-of-the-art methods for object detection existing in the literature.

About the Author

Peifang Guo:Postdoctoral researcher at University of Cincinnati in USA; Prabir Bhattacharya: Professor at University of Cincinnati in USA.