Complexity Estimation of Infrared Image Sequence for Automatic Target Track

Xiaotian Wang,Wanchao Ma,Kai Zhang,Shaoyi Li,Jie Yan

doi:10.1051/jnwpu/20193740664

Abstract

Infrared image complexity metrics are an important task of automatic target recognition and track performance assessment. Traditional metrics, such as statistical variance and signal-to-noise ratio, targeted to single frame infrared image. However, there are some studies on the complexity of infrared image sequences. For this problem, a method to measure the complexity of infrared image sequence for automatic target recognition and track is proposed. Firstly, based on the analysis of the factors affecting the target recognition and track, the specific reasons which background influences target recognition and track are clarified, and the method introduces the feature space into confusion degree of target and occultation degree of target respectively. Secondly, the feature selection is carried out by using the grey relational method, and the feature space is optimized, so that confusion degree of target and occultation degree of target are more reasonable, and statistical formula F1-Score is used to establish the relationship between the complexity of single-frame image and the two indexes. Finally, the complexity of image sequence is not a linear sum of the single-frame image complexity. Target recognition errors often occur in high-complexity images and the target of low-complexity images can be correctly recognized. So the neural network Sigmoid function is used to intensify the high-complexity weights and weaken the low-complexity weights for constructing the complexity of image sequence. The experimental results show that the present metric is more valid than the other, such as sequence correlation and inter-frame change degree, has a strong correlation with the automatic target track algorithm, and which is an effective complexity evaluation metric for image sequence.

Highlights

由图 7a) 度量指标散点图可知,306 组图像序列中跳出时刻目标遮隐度和目标相似度聚类中心为 (0.446,0.448) ,目标混淆度值都大于 0.3,且有一定遮隐度,造成目标跟踪过程中,无法准确识别目标, 跟踪虚假目标跳出视场。由图 7b) 复杂度分布可知,跳出时刻当帧复杂度值普遍较大,均值为 0.464。由实验可知,目标跟踪失败主要是由于跳出时刻复杂度较大引起的,而且目标相似度引入虚警的能力较大,也有一定目标遮隐度,该实验定性地分析了目标跳出视场的原因,在一定程度上,说明了在构建图像序列复杂度时,引入神经网络 Sigmoid 函数的合理性。由图 7e) 可知,在实现有效跟踪的 11 649 组序列中,5 个等级平均跟踪误差与本文提出的序列复杂度基本呈现单调递增的关系,在此基础上,与序列相关度和帧间目标变化度进行对比,如图 7c) 和 7d)所示, 本文序列复杂度指标单调性明显更好。该实验结果说明,本文序列复杂度的评价结果优于已有的 2 种评价标准。
[3] 侯旺, 梅风华, 陈国军,等. 基于背景最佳滤波尺度的红外图像复杂度评价准则[ J] . 物理学报, 2015, 64(23) : 95⁃104 HOU Wang, MEI Fenghua, CHEN Guojun, et al A Criterion for Evaluating the Complexity of Infrared Images Based on the Background Optimal Filtering Scale[ J]
Infrared image complexity metrics are an important task of automatic target recognition and track per⁃ formance assessment

Summary

Introduction

由图 7a) 度量指标散点图可知,306 组图像序列中跳出时刻目标遮隐度和目标相似度聚类中心为 (0.446,0.448) ,目标混淆度值都大于 0.3,且有一定遮隐度,造成目标跟踪过程中,无法准确识别目标, 跟踪虚假目标跳出视场。由图 7b) 复杂度分布可知,跳出时刻当帧复杂度值普遍较大,均值为 0.464。由实验可知,目标跟踪失败主要是由于跳出时刻复杂度较大引起的,而且目标相似度引入虚警的能力较大,也有一定目标遮隐度,该实验定性地分析了目标跳出视场的原因,在一定程度上,说明了在构建图像序列复杂度时,引入神经网络 Sigmoid 函数的合理性。由图 7e) 可知,在实现有效跟踪的 11 649 组序列中,5 个等级平均跟踪误差与本文提出的序列复杂度基本呈现单调递增的关系,在此基础上,与序列相关度和帧间目标变化度进行对比,如图 7c) 和 7d)所示, 本文序列复杂度指标单调性明显更好。该实验结果说明,本文序列复杂度的评价结果优于已有的 2 种评价标准。红外与激光工程, 2013, 42( 增刊 1) : 253⁃261 QIAO Liyong, XU Lixin, GAO Min. Influence of Infrared Image Complexity on the Target Detection Performance[ J] . 成都:电子科技大学, 2015 ZHENG Xin. Evaluation Method and Application Research of Infrared Image without Reference Map[ D] . 物理学报, 2015, 64(23) : 95⁃104 HOU Wang, MEI Fenghua, CHEN Guojun, et al A Criterion for Evaluating the Complexity of Infrared Images Based on the Background Optimal Filtering Scale[ J] .

Results

Conclusion