融合判别性与细粒度特征的抗遮挡红外目标跟踪算法  

Anti-Occlusion Infrared Target Tracking Algorithm Based on Fusion of Discriminant and Fine-Grained Features

作　　者：吴捷[1] 马小虎[2] WU Jie;MA Xiaohu(College of Information technology,Taizhou Polytechnic College,Taizhou 225300,China;School of Computer Science and Technology,Soochow University,Suzhou 215006,China)

机构地区：[1]泰州职业技术学院信息技术学院,江苏泰州225300 [2]苏州大学计算机科学与技术学院,江苏苏州215006

出　　处：《红外技术》2022年第11期1139-1145,共7页Infrared Technology

基　　金：国家自然科学基金(61402310);江苏省自然科学基金(BK20141195);泰州职业技术学院重点科研项目(1821819039)。

摘　　要：针对现有热红外目标跟踪算法难以处理相似物干扰和目标遮挡的问题,引入MMNet(Multi-task MatchingNetwork)算法中的多任务框架获取热红外目标特定的判别性特征和细粒度特征,并将这两种特征相互融合,用于在类间和类内识别热红外对象。此外,利用峰值旁瓣比动态设置模型更新参数以更高效地获取目标变化信息并对跟踪结果进行评估。对于不可靠跟踪结果利用卡尔曼滤波对目标位置进行预测。在LSOTB-TIR(Large-Scale Thermal Infrared Object Tracking Benchmark)红外数据集上的实验结果表明,提出的改进算法性能较好,相比MMNet跟踪精确度和成功率分别提高了5.7%和4.2%,且能有效应对遮挡、变形等挑战,可以应用于红外目标跟踪领域。Considering the problem in which the existing thermal infrared target tracking algorithms have difficulty dealing with similar object interference and target occlusion,the multi-task framework in the MMNet algorithm is introduced to obtain the specific discriminant features and fine-grained features of thermal infrared targets,which are fused to identify thermal infrared objects between and within classes.In addition,the peak side-lobe ratio is adopted to dynamically set the model update parameters and obtain the target change information more efficiently,in addition to evaluating the tracking results.For unreliable tracking results,a Kalman filter was unutilized to predict the target.The experimental results on the LSOTB-TIR dataset demonstrated that the performance of the improved algorithm was optimal.Compared with MMNet,the tracking accuracy and success rate were improved by 5.7%and 4.2%,respectively.It can effectively address the challenges of occlusion and deformation and can also be applied to the field of infrared target tracking.

关 键 词：热红外 判别性特征 细粒度特征 峰值旁瓣比 卡尔曼滤波 

分 类 号：TN911.73[电子电信—通信与信息系统]