主讲人：Chein-I Chang, Department of Computer Science and Electrical Engineering, University of Maryland, Baltimore County, Baltimore, USA

时间：2021年7月19日9：00-11：00(北京时间)

线下参与方式：机器人视觉感知与控制技术国家工程实验室 305报告厅集中听课

线上参与方式：

加入 Zoom 会议链接: https://us05web.zoom.us/j/87357345709?pwd=VUc1SUNmL3dqcndnSUlCeFVOZWtVUT09 

会议号：873 5734 5709 密码：123

主讲人简介: 

    Chein-I Chang 教授于1987年毕业于美国马里兰大学帕克分校，电机工程专业博士。现任美国马里兰大学巴尔地摩郡分校电机工程系终身教授，同时也是IEEE Life Fellow与SPIE Fellow、大连海事大学讲座教授、台湾中兴大学遥测科技杰出讲座教授，并担任IEEE Transaction on Geoscience and Remote Sensing、Remote Sensing等多个国际知名期刊编委，已发表SCI检索学术论文200余篇，其中超过100篇属于高光谱领域，撰写高光谱领域专著4部，并编著高光谱领域书籍3部，授权美国专利7项，Google被引次数达25000多次，Google学术H指数为66。

讲座简介: 

    Hyperspectral target detection can be performed in two different modes, active detection such as known target detection and passive detection such as anomaly detection. To evaluate detection performance, a general criterion is to use the area under a receiver operating characteristic (ROC) curve, AUC which is plotted based on detection probability, P_D versus false alarm probability, P_F. Unfortunately, Unfortunately, many ROC curves reported in the literature are indeed incorrectly generated. Another major issue is that using AUC of a ROC curve of (P_D,P_F), denoted by AUC_(D,F) is unreliable and misleading because P_D and P_F are generated by the same threshold. As a result, a higher P_D also generates a higher P_F and vice versa. To address these two issues this talk presents a 3D ROC analysis which generates a 3D ROC curve as a function of (P_D,P_F,?) by including the threshold parameter as a third independent variable. Consequently, a 3D ROC curve along with its derived three 2D ROC curves of (P_D,P_F), (P_D,?) and (P_F,?) can be further used to design new quantitative measures to evaluate the effectiveness of a detector and its target detectability TD and background suppressibility (BS). To demonstrate the full utility of 3D ROC analysis in target detection, examples are included in this talk to demonstrate how 3D ROC curves can be used to design new detection measures to evaluate target/anomaly detection performance more effectively ad accurately in terms, TD, BS and detector’s effectiveness.