为改善以人工为主的传统监控系统,在天津港煤码头无人区运用智能视频监控技术进行智能人形识别监控研究.该技术利用优化的方向梯度直方图(Histogram of Oriented Gradient,HOG)算法快速对人体轮廓进行描述;结合基于港口实际场景训练得...为改善以人工为主的传统监控系统,在天津港煤码头无人区运用智能视频监控技术进行智能人形识别监控研究.该技术利用优化的方向梯度直方图(Histogram of Oriented Gradient,HOG)算法快速对人体轮廓进行描述;结合基于港口实际场景训练得到的支持向量机分类器,标定出图像中有人的区域.在天津港煤码头无人区的现场实验表明,该技术对一幅320×240像素的图像的检测时间小于200 ms,满足港口监控实时性的要求.对具有复杂背景的监控区域,该技术能够高效地进行人形目标的匹配与识别,从而使安全得到更有效的保障.展开更多
鉴于目前散货码头运用智能视频监控系统时,由于不同方向人形的方向梯度直方图(Histogram of Oriented Gradient,HOG)特征存在较大的变化,使得用传统方法训练获得的少量特异性特征不足以支撑人形的有效分类,因此提出一种基于Ada Boost的...鉴于目前散货码头运用智能视频监控系统时,由于不同方向人形的方向梯度直方图(Histogram of Oriented Gradient,HOG)特征存在较大的变化,使得用传统方法训练获得的少量特异性特征不足以支撑人形的有效分类,因此提出一种基于Ada Boost的针对不同姿势HOG特征的二级分类方法.首先将样本快速分为正(背)面人形和侧面人形,组成第一级分类;然后通过分别为两类样本训练子分类器组成第二级分类;第二级分类对人形进行识别,并对结果进行融合.以天津港干散货码头无人作业区为背景,完成一组人形识别实验.实验结果表明,相较于传统方法,该方法对正(背)面人形具有更高的识别率.二级分类方法整体上提高了人形识别的识别率.展开更多
Based on the Fourier transform, a new shape descriptor was proposed to represent the flame image. By employing the shape descriptor as the input, the flame image recognition was studied by the methods of the artificia...Based on the Fourier transform, a new shape descriptor was proposed to represent the flame image. By employing the shape descriptor as the input, the flame image recognition was studied by the methods of the artificial neural network(ANN) and the support vector machine(SVM) respectively. And the recognition experiments were carried out by using flame image data sampled from an alumina rotary kiln to evaluate their effectiveness. The results show that the two recognition methods can achieve good results, which verify the effectiveness of the shape descriptor. The highest recognition rate is 88.83% for SVM and 87.38% for ANN, which means that the performance of the SVM is better than that of the ANN.展开更多
Basic block is the foundation of clothing construction design because it is the media between body and clothes and the fitness of clothes should be based on the accuracy of basic block. That needs us to recognize body...Basic block is the foundation of clothing construction design because it is the media between body and clothes and the fitness of clothes should be based on the accuracy of basic block. That needs us to recognize body not to record it. This paper reports the Algorithm of woman body fuzzy pattern recognition. It is organized in three sections:(i) extracting woman body feature; (ii) establishing membership functions of feature indexes;(iii) presenting an Algorithm for woman body fuzzy pattern recognition by example.展开更多
Accuracy and fastness of iris localization are very important in automatic iris recognition. A new fast iris localization algorithm based on improved generalized symmetry transform (GST) was proposed by utilizing (iri...Accuracy and fastness of iris localization are very important in automatic iris recognition. A new fast iris localization algorithm based on improved generalized symmetry transform (GST) was proposed by utilizing (iris) symmetry. GST was improved in three aspects:1) A new distance weight function is defined. The new weight function, which is effective in iris localization, utilized the characteristic of irises that the iris is a circular object and it has one inner boundary and one outer boundary. 2) Each calculation of the symmetry measurement of a pair of symmetry points was performed by taking one point of a pair as the starting point of the transformation. This is the most important reason for fast iris localization,due to which, repetitious computation was largely excluded. 3) A new phase weight function was proposed to adjust GST to locate circle target much better because the inner part of iris is darker than the outer part. The edge map of iris image was acquired and GST was only implemented on the edge point, which decreased computation without loss of accuracy. The modification of distance weight function and phase weight function leads to the accuracy of localization, and other ideas speed up the localization. Experiments show that the average speed of new algorithm is about 7.0—8.5 times as high as traditional ones including integro-differential operator and Hough transform method.展开更多
The facial expression recognition systn using the Ariaboost based on the Split Rectangle feature is proposed in this paper. This system provides more various featmes in increasing speed and accuracy than the Haarolike...The facial expression recognition systn using the Ariaboost based on the Split Rectangle feature is proposed in this paper. This system provides more various featmes in increasing speed and accuracy than the Haarolike featrue of Viola, which is commonly used for the Adaboost training algorithm. The Split Rectangle feature uses the nmsk-like shape composed with 2 independent rectangles, instead of using mask-like shape of Haar-like feature, which is composed of 2 --4 adhered rectangles of Viola. Split Rectangle feature has less di- verged operation than the Haar-like feaze. It also requires less oper- ation because the stun of pixels requires ordy two rectangles. Split Rectangle feature provides various and fast features to the Adaboost, which produrces the strong classifier with increased accuracy and speed. In the experiment, the system had 5.92 ms performance speed and 84 %--94 % accuracy by leaming 5 facial expressions, neutral, happiness, sadness, anger and surprise with the use of the Adaboost based on the Split Rectangle feature.展开更多
文摘为改善以人工为主的传统监控系统,在天津港煤码头无人区运用智能视频监控技术进行智能人形识别监控研究.该技术利用优化的方向梯度直方图(Histogram of Oriented Gradient,HOG)算法快速对人体轮廓进行描述;结合基于港口实际场景训练得到的支持向量机分类器,标定出图像中有人的区域.在天津港煤码头无人区的现场实验表明,该技术对一幅320×240像素的图像的检测时间小于200 ms,满足港口监控实时性的要求.对具有复杂背景的监控区域,该技术能够高效地进行人形目标的匹配与识别,从而使安全得到更有效的保障.
文摘鉴于目前散货码头运用智能视频监控系统时,由于不同方向人形的方向梯度直方图(Histogram of Oriented Gradient,HOG)特征存在较大的变化,使得用传统方法训练获得的少量特异性特征不足以支撑人形的有效分类,因此提出一种基于Ada Boost的针对不同姿势HOG特征的二级分类方法.首先将样本快速分为正(背)面人形和侧面人形,组成第一级分类;然后通过分别为两类样本训练子分类器组成第二级分类;第二级分类对人形进行识别,并对结果进行融合.以天津港干散货码头无人作业区为背景,完成一组人形识别实验.实验结果表明,相较于传统方法,该方法对正(背)面人形具有更高的识别率.二级分类方法整体上提高了人形识别的识别率.
基金Project(60634020) supported by the National Natural Science Foundation of China
文摘Based on the Fourier transform, a new shape descriptor was proposed to represent the flame image. By employing the shape descriptor as the input, the flame image recognition was studied by the methods of the artificial neural network(ANN) and the support vector machine(SVM) respectively. And the recognition experiments were carried out by using flame image data sampled from an alumina rotary kiln to evaluate their effectiveness. The results show that the two recognition methods can achieve good results, which verify the effectiveness of the shape descriptor. The highest recognition rate is 88.83% for SVM and 87.38% for ANN, which means that the performance of the SVM is better than that of the ANN.
文摘Basic block is the foundation of clothing construction design because it is the media between body and clothes and the fitness of clothes should be based on the accuracy of basic block. That needs us to recognize body not to record it. This paper reports the Algorithm of woman body fuzzy pattern recognition. It is organized in three sections:(i) extracting woman body feature; (ii) establishing membership functions of feature indexes;(iii) presenting an Algorithm for woman body fuzzy pattern recognition by example.
文摘Accuracy and fastness of iris localization are very important in automatic iris recognition. A new fast iris localization algorithm based on improved generalized symmetry transform (GST) was proposed by utilizing (iris) symmetry. GST was improved in three aspects:1) A new distance weight function is defined. The new weight function, which is effective in iris localization, utilized the characteristic of irises that the iris is a circular object and it has one inner boundary and one outer boundary. 2) Each calculation of the symmetry measurement of a pair of symmetry points was performed by taking one point of a pair as the starting point of the transformation. This is the most important reason for fast iris localization,due to which, repetitious computation was largely excluded. 3) A new phase weight function was proposed to adjust GST to locate circle target much better because the inner part of iris is darker than the outer part. The edge map of iris image was acquired and GST was only implemented on the edge point, which decreased computation without loss of accuracy. The modification of distance weight function and phase weight function leads to the accuracy of localization, and other ideas speed up the localization. Experiments show that the average speed of new algorithm is about 7.0—8.5 times as high as traditional ones including integro-differential operator and Hough transform method.
基金supported by the Brain Korea 21 Project in2010,the MKE(The Ministry of Knowledge Economy),Koreathe ITRC(Information Technology Research Center)support programsupervised by the NIPA(National ITIndustry Promotion Agency)(NI-PA-2010-(C1090-1021-0010))
文摘The facial expression recognition systn using the Ariaboost based on the Split Rectangle feature is proposed in this paper. This system provides more various featmes in increasing speed and accuracy than the Haarolike featrue of Viola, which is commonly used for the Adaboost training algorithm. The Split Rectangle feature uses the nmsk-like shape composed with 2 independent rectangles, instead of using mask-like shape of Haar-like feature, which is composed of 2 --4 adhered rectangles of Viola. Split Rectangle feature has less di- verged operation than the Haar-like feaze. It also requires less oper- ation because the stun of pixels requires ordy two rectangles. Split Rectangle feature provides various and fast features to the Adaboost, which produrces the strong classifier with increased accuracy and speed. In the experiment, the system had 5.92 ms performance speed and 84 %--94 % accuracy by leaming 5 facial expressions, neutral, happiness, sadness, anger and surprise with the use of the Adaboost based on the Split Rectangle feature.
