检索结果-内蒙古大学图书馆

International Conference on Virtual Reality and Visualization (ICVRV)

作者： Guo, Wei Xiao, Ya Zhang, Guodong Shenyang Aerosp Univ Sch Comp Shenyang 110136 Peoples R China

ISBN: (纸本)9781479968541

Pedestrian detection has a wide range of applications in visual surveillance, driver assistance systems. It is also very important in computer vision and pattern recognition. In our study, we proposed a multi-scale scheme for pedestrian detection. The scheme of pedestrian detection consisted of two steps for construction a strong classifier and multi-scale detection. The strong classifier, a collection of weak classifiers, was built by use of adaboost learning algorithm based on the Harr-like features. Then, the strong classifier was employed to detect pedestrians in the multi-scale images, and the detection results were merged. In our experiment, the proposed multi-scale detection scheme reported 0.35 false positives per image at the sensitivity to 89.3%. This indicates that the multi-scale scheme for pedestrian detection achieves a high performance.

关键词： Haar-like features adaboost learning algorithm Multi-scale detection

来源：评论

学校读者我要写书评

暂无评论

Ensemble of adaboost cascades of 3L-LBPs classifiers for license plates detection with low quality images

引用

EXPERT SYSTEMS WITH APPLICATIONS 2018年 92卷 216-235页

作者： Al-Shemarry, Meeras Salman Li, Yan Abdulla, Shahab Univ Southern Queensland Fac Hlth Engn & Sci Sch Agr Computat & Environm Sci Toowoomba Qld Australia Univ Southern Queensland Open Access Coll Toowoomba Qld 4350 Australia

Due to the plate formats and multiform outdoor illumination conditions during the image acquisition phase, it is challenging to find effective license plate detection (LPD) method. This paper aims to develop a new detection method for identifying vehicle license plates under low quality images using image processing techniques. In this research, a robust method using a large number of adaboost cascades with three levels pre-processing local binary patterns classifiers (3L-LBPs) are used to detect license plates (LPs) regions. The method achieves a very high accuracy for detecting LP number from one vehicle image. The proposed method was tested and trained with the images from 630 and 400 vehicles, respectively. The images involve many difficult conditions, such as low/high contrast, dusk, dirt, fogy, and distortion problems. The experimental results demonstrate very satisfactory performance for LP detection in term of speed and accuracy, and were better than the most of the existing methods. The processing time for the whole testing LPD system was about 1.63 seconds to 2 seconds. The overall probability detection, precision, and f-measurement are 98.56%, 95.9% and 97.19%, respectively;with false positive rate 5.6%. (C) 2017 Elsevier Ltd. All rights reserved.

关键词： License plate detection (LPD) Region of interest (ROl) adaboost learning algorithm Cascade classifier Local binary pattern classifiers (LBP)

来源：评论

学校读者我要写书评

暂无评论

3D face pose estimation by a robust real time tracking of facial features

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2016年第23期75卷 15693-15708页

作者： Chun, Junchul Kim, Wonggi Kyonggi Univ Dept Comp Sci San 94-6 Yiui Dong Suwon South Korea

In this paper, we present a new 3D head pose estimating approach based on real-time facial feature tracking and facial feature recovering method which copes with the surrounding light variation and various occlusion. The major facial features are obtained by Haar-like feature detection along with adaboost learning from an input video image. The detected facial features are robustly tracked by optical flow with a template matching scheme which continuously compensates for losing track of the initially detected features in a sequence of input images. The head pose of an input face image can be obtained by evaluating 3D information of facial features from the detected 2D eye-points, nose and lip. From the experiments, the proposed method shows effectiveness in tracking and recovering facial features and produces reliable result in head pose estimation.

关键词： Head pose estimation Haar-like feature detection Optical flow adaboost learning algorithm Template matching

来源：评论

学校读者我要写书评

暂无评论

Automated Blog Feedback Prediction with Ada-Boost Classifier 4

Automated Blog Feedback Prediction with Ada-Boost Classifier

引用

4th International Conference on Informatics, Electronics & Vision (lCIEV)

作者： Uddin, Md. Taufeeq Int Islamic Univ Chittagong Dept Comp Sci & Engn Chittagong Bangladesh

ISBN: (纸本)9781467369022

Automated analysis of social media documents has a tremendous impact in our day to day life since we extensively use social media to share our thoughts, feelings, tastes etc. However, the automatic social media analysis is still a very challenging task due to the massive amount of social media documents as well as the uncontrolled, dynamic and rapidly-changing content of social media documents. To automate social media analysis, this paper presents an automatic feedback prediction model based on novel Ada-Boost learning algorithm for blog documents considering realistic scenario. In this approach, an Ada-Boost classifier is applied to the numerous features extracted from crawled blog document to predict whether someone comments on a blog document or not in the next 24 hours of its publication in blogs. The evaluation results of the experiments conducted on the publicly available benchmark blog feedback data set indicate that the proposed technique is efficient both in terms of feedback prediction accuracy and computational time;the proposed approach yielded the maximum feedback prediction rate of 91.4%.

关键词： classification document handling electronic publishing feature extraction learning (artificial intelligence) pattern classification social networking (online) adaboost classifier adaboost learning algorithm automated blog feedback prediction automated social media documents analysis benchmark blog feedback data set blog document computational time feedback prediction accuracy Blogs Data mining Feature extraction Media Prediction algorithms Predictive models Training Feature extraction blogs electronic document handling Electronic Publishing Social Media Prediction algorithms computational time Pattern recognition Prediction models learning algorithms data mining Functional training Feedback

来源：评论

学校读者我要写书评

暂无评论

Computerized Scheme for Vertebra Detection in CT scout image

Computerized Scheme for Vertebra Detection in CT scout image

引用

Conference on Medical Imaging - Computer-Aided Diagnosis

作者： Guo, Wei Chen, Qiang Zhou, Hanxun Zhang, Guodong Cong, Lin Li, Qiang Shenyang Aerosp Univ Sch Comp Shenyang 110136 Liaoning Peoples R China Liaoning Univ Dept Informat Sci Shenyang 110036 Liaoning Peoples R China Shanghai United Imaging Healthcare Co Ltd Shanghai 201815 Peoples R China Chinese Acad Sci Shanghai Adv Res Inst Ctr Adv Med Imaging Technol Shanghai 201210 Peoples R China

ISBN: (纸本)9781510600201

Our purposes are to develop a vertebra detection scheme for automated scan planning, which would assist radiological technologists in their routine work for the imaging of vertebrae. Because the orientations of vertebrae were various, and the Haar-like features were only employed to represent the subject on the vertical, horizontal, or diagonal directions, we rotated the CT scout image seven times to make the vertebrae roughly horizontal in least one of the rotated images. Then, we employed adaboost learning algorithm to construct a strong classifier for the vertebra detection by use of Haar-like features, and combined the detection results with the overlapping region according to the number of times they were detected. Finally, most of the false positives were removed by use of the contextual relationship between them. The detection scheme was evaluated on a database with 76 CT scout image. Our detection scheme reported 1.65 false positives per image at a sensitivity of 94.3% for initial detection of vertebral candidates, and then the performance of detection was improved to 0.95 false positives per image at a sensitivity of 98.6% for the further steps of false positive reduction. The proposed scheme achieved a high performance for the detection of vertebrae with different orientations.

关键词： Automated scan planning vertebra detection adaboost learning algorithm contextual relationship

来源：评论

学校读者我要写书评

暂无评论

Channel estimation and symbol detection for OFDM systems using data-nulling superimposed pilots

引用

ELECTRONICS LETTERS 2014年第3期50卷 179-180页

作者： Dou, Gaoqi He, Chunquan Li, Congying Gao, Jun Naval Univ Engn Dept Commun Engn Wuhan Peoples R China

A novel data-nulling superimposed pilot scheme for orthogonal frequency division multiplexing (OFDM) systems is proposed, where the input data vector is spread over all the subcarriers by a precoding matrix and then nulled at certain subcarriers for the insertion of training pilots. This method avoids the loss of the data rate for frequency-division multiplexed pilots, but results in the distortion of input data. To mitigate the distortion introduced by the nulling operation, a simple iterative reconstruction scheme is used to improve the detection performance.

关键词： geometry learning (artificial intelligence) object recognition geometric features 3D range data three-dimensional range data tree recognition human recognition off-road environment adaboost learning algorithm discriminative features

来源：评论

学校读者我要写书评

暂无评论

Hybrid Method for Hand Gesture Recognition Based on Combination of Haar-Like and HOG Features

Hybrid Method for Hand Gesture Recognition Based on Combinat...

引用

21st Iranian Conference on Electrical Engineering (ICEE)

作者： Ghafouri, Sudabeh Seyedarabi, Hadi Univ Coll Nabi Akram Tabriz Iran Univ Tabriz Fac Elect & Comp Engn Tabriz Iran

ISBN: (纸本)9781467356343

In this paper a new method is proposed for hand gesture recognition. The proposed method increases hand gesture recognition rate and decreases false positive error rate by using combination of Haar-like and Histogram of Oriented Gradients (HOG) features. Also some new Haar-like features are proposed proportional to hand posture to solve major Haar-like problem that is high false positive error rate in hand posture recognition. These features improve recognition rate to 83%. The experiments showed that hybrid method can recognize hand gesture by 93.5% accuracy which is 25% higher than previous method, and decrease the false positive error from 92% to 8%.

关键词： Hand posture recognition Haar-like feature adaboost learning algorithm Histogram Gradient Oriented feature Multi-Class Support Vector Machine

来源：评论

学校读者我要写书评

暂无评论

Automatic Nipple Detection Using Cascaded adaboost Classifier

Automatic Nipple Detection Using Cascaded AdaBoost Classifie...

引用

5th International Symposium on Computational Intelligence and Design (ISCID)

作者： Xin Kejun Wu Jian Ni Pengyu Huang Jie Nanjing Sampel Technol Co Ltd Nanjing Jiangsu Peoples R China Southeast Univ Sch Informat Sci & Engn Nanjing Jiangsu Peoples R China

ISBN: (纸本)9780769548111

Many non-pornographic images containing large exposure of skin area or approximate skin-color area are prone to be detected as the pornographic images. This paper proposes a novel method based on adaboost algorithm for nipple detection of pornographic images. The adaboost algorithm has excellent performance in both detection accuracy and detection speed. The method extracts extended Haar-like features, color features, texture features and shape features to train and obtain a cascaded adaboost classifier by using adaboost algorithm. And it is validated for locating nipple existence in pornographic images. The experimental results show that this method performs well for nipple detection in pornographic images, and can reduce effectively the false positive rate against the non-pornographic images.

关键词： adaboost learning algorithm Haar-like feature color features texture features shape features cascaded adaboost classifier

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：