自动目标识别中的图像序列质量评价方法

刁伟鹤; 毛峡; 常乐

doi:10.3724/SP.J.1146.2009.01194

自动目标识别中的图像序列质量评价方法

doi: 10.3724/SP.J.1146.2009.01194

基金项目:

航天支撑技术基金和北京航空航天大学博士生创新基金资助课题

计量
- 文章访问数: 4106
- HTML全文浏览量: 70
- PDF下载量: 1078
- 被引次数: 0
出版历程
- 收稿日期: 2009-09-08
- 修回日期: 2009-11-26
- 刊出日期: 2010-08-19

Quality Estimation of Image Sequence for Automatic Target Recognition

摘要

摘要: 图像质量评价是自动目标识别(ATR)性能评估中的重要组成部分。传统的评价指标如目标信噪比(SNR)等针对的皆是单幅图像，而对于图像序列质量评价的研究尚属空白。针对该问题，该文首次提出了帧间目标变化程度的概念，利用其定量描述图像序列的质量。该指标融合了3部分信息：图像序列中帧间目标区纹理变化的信息，帧间目标大小变化的信息以及帧间目标位置不规律的信息。为验证所提指标的有效性，设计了用于分析帧间目标变化程度和ATR算法实际效果关系的实验，实验用样本为真实的目标图像序列。实验结果表明，帧间目标变化程度与ATR算法性能具有很强的相关性，基本呈现单调关系，是一种有效的图像序列质量评价标准。
- 自动目标识别性能评估 /
- 图像序列质量评价 /
- 帧间目标变化程度
Abstract: Image quality estimation is an important part of performance evaluation for Automatic Target Recognition (ATR). Traditional image metrics are focused on single image, and there is no valid method to estimate the quality of image sequence. According to the above problem, this paper proposes the concept Inter-Frame Change Degree of Target (IFCDT) which is used for quantitatively describing image sequence quality for the first time. There are three key elements in the formula of the proposed metric, which embody the information of inter-frame change of target image texture, inter-frame change of target size and inter-frame change of target position respectively. To validate this image sequence metric, this paper designs an experiment for analyzing the relationship of IFCDT and ATR algorithm performance, while the samples of this experiment are actual target image sequences. The experiment result shows that there is a good monotone relationship between the image sequence metric and algorithm actual performance. Therefore, it can be concluded that the metric is a valid criterion to evaluate the quality of target image sequence.

HTML全文

参考文献(1)

Li Min and Zhang Gui-lin. Image measures for segmentationalgorithm evaluation of automatic target recognitionsystem[C]. 1st International Symposium on Systems andControl in Aerospace and Astronautics, Harbin, China, 2006:674-679.[2]Edmondson R, Rodgers M, and Banish M, et al.. Single-frameimage processing techniques for low-SNR infraredimagery[C]. Proceeding of SPIE, 2008, 6940: 74-78.[3]Yang L, Zhou Y, Yang J, and Chen L. Variance WIE basedinfrared images processing[J].Electronics Letters.2006,42(15):857-859[4]Clark L G and Vincent V J. Image characterization forautomatic target recognition algorithm evaluations[J].Optical Engineering.1991, 30(2):147-153[5]Trievdi M M and Schirvaikar M V. Quantitativecharacterization of image clutter: problems, progress, andpromises[C][J].proceedings of SPIE.1993, 1967:288-299[6]李敏, 周振华, 张桂林. 自动目标识别算法性能评估中的图像度量研究[J]. 红外与激光工程, 2007, 36(3): 412-416.Li Min, Zhou Zhen-hua, and Zhang Gui-lin. Image measuresin the evaluation of ATR algorithm performance [J]. Infraredand Laser Engineering, 2007, 36(3): 412-416.[7]周川, 张桂林, 陈鸿翔等. 基于试验设计的ATR 算法的性能评估[J]. 华中理工大学学报, 1996, 24(2): 43-45.Zhou Chuan, Zhang Gui-lin, and Chen Hong-xiang.Performance evaluation for ATR algorithms based onexperiments design[J]. Journal of Huazhong University ofScience and Technology, 1996, 24(2): 43-45.[8]Mao Xia and Diao Wei-he. Criterion to evaluate the qualityof infrared small target images[J].Journal of Infrared,Millimeter, and Terahertz Waves.2009, 30(1):56-64[9]常洪花, 张建奇. 基于人眼视觉的红外背景杂波量化技术[J].红外技术, 2004, 26(5): 13-18.Chang Hong-hua and Zhang Jian-qi. Human vision-based onquantitative characterization of IR background clutter [J].Infrared Technology, 2004, 26(5): 13-18.[10]Chang Hong-hua and Zhang Jian-qi. Evaluation of humandetection performance using target structure similarityclutter metrics[J]. Optical Engineering, 2006, 45(9): 41-47.[11]Aviram G and Rotman S R. Analyzing the effect of imagerywavelength on the agreement between various image metricsand human detection performance of targets embedded innatural images[J].Optical Engineering.2008, 40(9):1877-1884[12]He Guo-jing, Zhang Jian-qi, and Chang hong-hua. Cluttermetric based on the Cramer-Rao lower bound on automatictarget recognition[J].Applied Optics.2008, 47(29):5534-5540[13]Rotman S R, Hsu D, Cohen A, Shamay D, and Kowalczyk M.Textural metrics for clutter affecting human targetacquisition[J].Infrared Physics Technology.1996, 37(6):667-674[14]Salem Y B and Nasri S. Texture classification of woven fabricbased on a glcm method and using multiclass support vectormachine[C]. 6th International Multi-Conference on Systems,Signals and Devices, Djerba Tunisia, 2009: 1-8.[15]Wang Jian-hui, Li Feng, Doi Kunio, and Li Qiang. A novelscheme for detection of diffuse lung disease in MDCT by useof statistical texture features[C]. Proceeding of SPIE, 2009,7260: 382-389.[16]杨磊. 复杂背景条件下的红外小目标检测与跟踪算法研究[D].[博士论文], 上海：上海交通大学图像处理与模式识别研究所,2006.Yang Lei. Study on infrared small target detection andtracking algorithm under complex backgrounds [D]. [Ph.D.dissertation], Shanghai: Institude of Image Processing andPattern Recognition, Shanghai Jiao Tong University, 2006.

施引文献

资源附件(0)

访问统计