基于结构张量降维和改进上下文感知相关滤波器的高光谱目标跟踪  

Hyperspectral Target Tracking Based on Dimensionality Reduction of Structural Tensors and Improved Context-Aware Correlation Filter

作　　者：赵东 胡斌 庄宇辰 滕翔 王超[1] 李佳[4] 郭业才[1,2] Zhao Dong;Hu Bin;Zhuang Yuchen;Teng Xiang;Wang Chao;Li Jia;Guo Yecai(School of Electronics&Information Engineering,Nanjing University of Information Science&Technology,Nanjing 210044,Jiangsu,China;School of Electronics and Information Engineering,Wuxi University,Wuxi 214105,Jiangsu,China;School of Physics,Xidian University,Xian 710071,Shaanxi,China;Department of Basic Sciences,Air Force Engineering University,Xi an 710051,Shaanxi,China)

机构地区：[1]南京信息工程大学电子与信息工程学院,江苏南京210044 [2]无锡学院电子与信息工程学院,江苏无锡214105 [3]西安电子科技大学物理学院,陕西西安710071 [4]中国人民解放军空军工程大学基础部,陕西西安710051

出　　处：《光学学报》2024年第11期150-160,共11页Acta Optica Sinica

基　　金：国家自然科学基金(62001443,62105258);江苏省自然科学基金(BK20210064);无锡市创新创业资金“太湖之光”科技攻关计划(基础研究)(K20221046);无锡学院人才启动基金(2021r007)。

摘　　要：针对高光谱目标跟踪算法在目标受遮挡时跟踪漂移的问题,本文提出一种基于结构张量降维和改进上下文感知相关滤波器的高光谱目标跟踪算法。首先利用结构张量提取目标区域和搜索区域的边缘纹理信息,再对结构张量进行分解得到对应目标区域和搜索区域的特征向量,通过计算每个波段目标区域特征向量与搜索区域特征向量之间的马氏距离,获得当前帧高光谱图像的多维光谱权重,再利用多维光谱权重与高光谱图像进行加权融合实现降维,同时利用光谱信息与VGG19网络来提取降维后图像的深度特征;然后在训练分类器时,为抑制循环位移带来的边界效应,通过计算响应图的干扰因子来改进引入的上下文信息,当后续帧因干扰因素导致跟踪产生误差,并且随时间增加导致累积误差超过设定阈值的时候,本文所提跟踪算法将初始帧的响应图与当前帧的响应图融合,以便及时对跟踪结果进行校正;最后在标准数据集上验证了算法的性能,实验结果表明,本文所提跟踪算法与对比算法相比较,在克服目标遮挡方面具有更好的鲁棒性。Objective Hyperspectral videos(HSVs)contain abundant spectral information to facilitate the capture of distinctive spectral characteristics of the target.In RGB images,traditional tracking algorithms are prone to failure when confronted with targets that share similar shape,size,or color with the background,or low spatial resolution.Hyperspectral images provide detailed information about the internal structure and chemical composition of the target in the form of a threedimensional data cube,where each target possesses a unique spectral curve.However,as the number of bands increases in hyperspectral images,both data complexity and computational complexity escalate,with diminishing data processing efficiency.Therefore,effective data compression becomes crucial.The occlusion problem frequently affects tracking accuracy and impedes real-time tracking implementation of target tracking tasks.Consequently,we aim to address challenges related to data processing and occlusion in hyperspectral target tracking by providing an efficient algorithm for reducing spectral matching discrepancies and suppressing tracking drift.Methods The algorithm is based on the context filter framework and incorporates the scale filter from the DSST algorithm as the scale estimation module.By computing the structure tensors of both the target and search regions,we extract edge structure features,reconstruct their respective structure tensors,and decompose them to obtain feature roots and corresponding feature vectors.By calculating the Mahalanobis distance between the target region and background region,we derive a multi-dimensional spectral weight which is then multiplied with the structure tensor of the search region.Finally,we calculate the Euclidean distance to achieve dimensionality reduction to bring about an image that is copied into three channels and inputted into the VGG19 network for extracting depth features.These features are subsequently fed into an enhanced context filter which improves upon traditional methods by enhancing cycl

关 键 词：目标跟踪 结构张量降维 光谱信息 干扰因子 上下文信息 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置] TP79[自动化与计算机技术—控制科学与工程]