作者：王驰1,2, 占李黎1, 于明坤1, 李富迪1, 张鼎2
作者单位：1. 上海大学精密机械工程系，上海 200444;
2. 近地面探测技术重点实验室，江苏 无锡 214035
关键词：抛撒地雷;地雷探测;微光夜视;机器学习;多雷区探测
摘要：
利用基于机器学习的抛撒地雷智能探测方法及探测系统，在夜间复杂场景下对不同类型的地雷目标进行智能探测实验研究。在论述抛撒地雷夜视智能识别方法的基础上，以72式防坦克金属地雷、69式防坦克塑壳地雷、58式防步兵橡胶地雷作为抛撒地雷目标，设计不同的实验场景进行被动式夜视智能探测实验，并研究多雷区的探测方法及可行性。结果显示，在给定的实验条件下，抛撒地雷的裸露面积由无明显遮挡、轻微遮挡至部分遮挡的检测精确率分别为98.97%、98.5%、87.5%，召回率分别为99.22%、71.3%、56.8%；在夜间对布设的多雷区进行探测，可有效排除孤立雷并实现对多雷区的标定。表明研究的探雷方法能实现对不同场景下抛撒地雷的智能识别，可用于复杂雷区夜视智能探测技术的进一步研究。
Research on night vision intelligent recognition method of scatterable landmines based on machine learning
WANG Chi1,2, ZHAN Lili1, YU Mingkun1, LI Fudi1, ZHANG Ding2
1. Department of Precision Mechanical Engineering, Shanghai University, Shanghai 200444, China;
2. Science and Technology on Near-surface Detection Laboratory, Wuxi 214035, China
Abstract: Using machine learning-based intelligent detection methods and systems for scatterable landmines, intelligent detection experiments on different types of landmine targets in complex night scenes are carried out. On the basis of discussing the intelligent identification method for the night vision of scatterable landmines, the 72 type anti-tank metal mines, the 69 type anti-tank plastic shell mines, and the 58 type anti-infantry rubber mines are used as the targets of the scatterable landmines, and different experimental scenes are designed for passive night vision intelligence. The minefield detection method and its feasibility are studied. The results show that under the given experimental conditions, the detection accuracy rates of the exposed area of landmines from no obvious occlusion, slight occlusion to partial occlusion are 98.97%, 98.5%, 87.5% and the recall rates are 99.22%, 71.3%, 56.8%. Detecting the deployed minefield at night can effectively eliminate isolated mines and realize the calibration of the minefield. It shows that the mine detection method studied can effectively realize the intelligent detection of the scatterable mines in different scenarios, and can be used for further research of the night vision intelligent detection of complex minefields.
Keywords: scatterable landmines;landmine detection;low-light night vision;machine learning;minefield detection
2022, 48(11):34-40 收稿日期: 2021-06-28;收到修改稿日期: 2021-09-07
基金项目: 国家自然科学基金（61773249）；近地面探测技术重点实验室基金（TCGZ2020C003）；上海市科技创新行动计划（20142200100）
作者简介: 王驰（1982-），男，河南太康县人，教授，博士，研究方向为应用光学与融合传感技术
参考文献
[1] BAUR J, STEINBERG G, NIKULIN A, et al. Applying deep learning to automate uav-based detection of scatterable landmines[J]. Remote Sensing, 2020, 12(5): 859-875
[2] YU Y, LIU B, CHEN Z, et al. A Macro-pulse photon counting lidar for long-range high-speed moving target detection[J]. Sensors, 2020, 20(8): 2204-2216
[3] 张石磊, 崔宇, 邢慕增, 等. 光场成像目标测距技术[J]. 中国光学, 2020, 13(6): 1332-1342
[4] FRIGUI H, ZHANG L, GADER P D. Context-dependent multisensor fusion and its application to land mine detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(6): 2528-2543
[5] KASBAN H, ZAHRAN O, SAYED M, et al. A comparative study of landmine detection techniques[J]. Sensing and Imaging:An International Journal, 2010, 11(3): 89-112
[6] 田文忠, 赵庆展, 胡浩伟, 等. 无人机高光谱载荷性能交叉验证[J]. 中国测试, 2019, 45(11): 131-137
[7] 何微微, 武魁军, 傅頔, 等. 临近空间风温遥感干涉仪设计及正演[J]. 光学精密工程, 2020, 28(8): 1678-1689
[8] MONTIEL-ZAFRA V, CANADAS-QUESADA F J, VERA-CANDEAS P, et al. A novel method to remove GPR background noise based on the similarity of non-neighboring regions[J]. Journal of Applied Geophysics, 2017, 144: 188-203
[9] 刘祚时, 周继雯, 俞跃, 等. 油气管道聚乙烯层粘接缺陷的红外热成像检测方法和信号增强技术研究[J]. 中国测试, 2019, 45(11): 1-8
[10] ŠIPOš D, GLEICH D. A Lightweight and low-power uav-borne ground penetrating radar design for landmine detection[J]. Sensors, 2020, 20(8): 2234-2252
[11] FREELAND R S, MILLER M L, YODER R E, et al. Forensic application of FM-CW and pulse radar[J]. Journal of Environmental and Engineering Geophysics, 2003, 8(2): 97-103
[12] 凡丽梅, 董方旭, 安志武, 等. 橡胶/铝合金粘接构件脱粘缺陷非线性超声检测技术研究[J]. 中国测试, 2020, 46(8): 15-21
[13] CHRZANOWSKI K. Review of night vision technology[J]. Opto-Electronics Review, 2013, 21(2): 153-181
[14] 刘超, 张晓晖. 超低照度下微光图像的深度卷积自编码网络复原[J]. 光学精密工程, 2018, 26(4): 951-961
[15] GADE R, MOESLUND T B. Thermal cameras and applications: a survey[J]. Machine Vision and Applications, 2014, 25(1): 245-262
[16] YUAN T, HAN Z H, LI L, et al. Tunable-liquid-crystal- filter-based low-light-level color night vision system and its image processing method[J]. Applied optics, 2019, 58(18): 4947-4955
[17] 王驰, 于明坤, 杨辰烨, 等. 抛撒地雷的夜视智能探测方法研究[J]. 中国光学, 2021, 14(5): 1202-1211.
