Facial/License Plate Detection Using a Two-Level Cascade Classifier and a Single Convolutional Feature Map

NCUIR > College of Electrical Engineering & Computer Science > Department of Computer Science and information Engineering > journal & Dissertation > Item 987654321/106780

Please use this identifier to cite or link to this item: https://ir.lib.ncu.edu.tw/handle/987654321/106780

Title:	Facial/License Plate Detection Using a Two-Level Cascade Classifier and a Single Convolutional Feature Map
Authors:	范國清;Chen, Ying-Nong;Han, Chin-Chuan;Ho, Gang-Feng;Fan, Kuo-Chin
Contributors:	資訊電機學院資訊工程學系
Keywords:	Algorithms;Automobile license plates;Classifiers;Convolution;Crime prevention;Face recognition;Feature maps;Identification and classification;Machine learning;Machine vision;Mappings (Mathematics);Methods;Neural networks;Principal components analysis;Ranking;Sensors;Training;Wavelet transforms
Date:	2015-12-18
Issue Date:	2026-04-23 13:42:05 (UTC+8)
Publisher:	SAGE Publications Inc.;London, England: SAGE Publications
Abstract:	摘要： In this paper, an object detector is proposed based on a convolution/subsampling feature map and a two-level cascade classifier. First, a convolution/subsampling operation alleviates illumination, rotation and noise variances. Then, two classifiers are concatenated to check a large number of windows using a coarse-to-fine strategy. Since the sub-sampled feature map with enhanced pixels was fed into the coarse-level classifier, the checked windows were drastically reduced to a quarter of the original image. A few remaining windows showing detailed data were further checked using a fine-level classifier. In addition to improving the detection process, the proposed mechanism also sped up the training process. Some features generated from the prototypes within the small window were selected and trained to obtain the coarse-level classifier. Moreover, a feature ranking algorithm reduced the large feature pool to a small set, thus speeding up the training process without losing detection performance. The contribution of this paper is twofold: first, the coarse-to-fine scheme shortens both the training and detection processes. Second, the feature ranking algorithm reduces training time. Finally, some experimental results were achieved for evaluation. From the results, the proposed method was shown to outperform the rapidly performing Adaboost, as well as forward feature selection methods. 出版者： London, England: SAGE Publications 出版日期： 2015-12-18 出處： International Journal of Advanced Robotic Systems, 2015-12, Vol.12 (12), p.1 資源來源： Openly Available Collection - Sage Journals 版權： 2015 Author(s). Licensee InTech. 版權： COPYRIGHT 2015 Sage Publications Ltd. (UK) 版權： 2015. This work is published under http://creativecommons.org/licenses/by/3.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. 識別號： ISSN: 1729-8806 識別號： ISSN: 1729-8814 識別號： EISSN: 1729-8814 識別號： DOI: 10.5772/61477
Appears in Collections:	[Department of Computer Science and information Engineering] journal & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	13	View/Open

社群 sharing

Loading...