深度迁移学习预训练对红外尾流成像识别的影响

钟睿; 杨立; 杜永成

深度迁移学习预训练对红外尾流成像识别的影响

海军工程大学动力工程学院，湖北武汉 430033

基金项目: "十三五"海军预研项目

详细信息

作者简介:
钟睿（1996-），男，硕士，主要研究方向：传热、热流体及其应用。E-mail：243225679@qq.com

通讯作者:
杨立（1962-），男，教授，主要研究方向：传热、热流体及其应用

中图分类号: TP391.41
计量
- 文章访问数: 108
- HTML全文浏览量: 62
- PDF下载量: 22
- 被引次数: 0
出版历程
- 收稿日期: 2020-03-08
- 修回日期: 2020-04-27
- 刊出日期: 2021-10-20

The Influence of Deep Transfer Learning Pre-training on Infrared Wake Image Recognition

School of Power Engineering, Naval University of Engineering, Wuhan 430033, China

摘要

摘要: 随着水下航行器噪声水平的不断降低，水下航行器形成的尾流红外成像特征就成为其主要可探测的特征源之一，利用水下航行器尾流的水面红外特征来探测水下航行器的踪迹逐渐发展成为一种新的探测方式。由于人工判别尾流特征的效率低，准确性不高，采用人工智能深度学习的方式能够得到较大的改善。本文以水下航行器尾流红外特征识别为研究核心，通过图像分类制作了混合类的样本集，利用迁移学习比较不同预训练网的对尾流的训练效果，讨论预训练网内部参数对尾流训练效果的影响，结合Faster-RCNN算法，最终测试对尾流的识别精度，在45个2类尾流的小样本集下，预训练之后的网络在识别准确度上增加了21.43%，误检率下降了2.14%，带有红外特征的图像在定位精准率上比可见光图像高18.18%。该预训练测试对未来研究尾流探测结合卷积神经网络的识别有一定的应用潜力。
- 红外特征 /
- 尾流 /
- 深度学习 /
- 迁移学习 /
- faster-RCN
Abstract: With lower underwater vehicle noise levels, the infrared imaging characteristics of underwater vehicle wake have become one of the main detectable sources. Using the infrared characteristics of underwater vehicle wakes to detect underwater vehicle traces has gradually developed into a popular detection method. Because of the low efficiency and inaccuracy of artificial wake characteristics identification, the adopted artificial intelligence deep learning method can be greatly improved. In this study, the infrared feature recognition of underwater vehicle wake is the primary focus. A sample set of mixed classes was made by image classification. The training effect of different pre-training networks was compared using migration learning. The influence of the internal parameters of the pre-training networks on the training effect of the wake was discussed. Finally, in the small sample set of 45 two kinds of wake, the recognition accuracy of the network after pre-training increased by 21.43%, the false detection rate decreased by 2.14%, and the positioning accuracy of the image with infrared characteristics was 18.18% higher than that of the visible image. This pre-training test has a certain application potential for future research on wake detection combined with convolution neural network recognition.
- infrared characteristics /
- wake /
- deep learning /
- transfer learning /
- faster-RCNN

HTML全文

图 1 不同类型的尾流特征图像

Figure 1. Different types of wake feature images

下载: 全尺寸图片幻灯片

图 2 Faster-RCNN在识别红外尾流中的应用

Figure 2. Application of faster-RCNN in infrared wake identification

下载: 全尺寸图片幻灯片

图 3 Google、VGG19、AlexNet基本网络对比实验图

Figure 3. Comparison experiment of Google, VGG 19 and AlexNet

下载: 全尺寸图片幻灯片

图 4 Frequency参数对AlexNet网络的影响实验

Figure 4. Experiment on the influence of frequency parameters on AlexNet network

下载: 全尺寸图片幻灯片

图 5 Patience参数对AlexNet网络影响实验

Figure 5. Experiment of influence of patience parameters on AlexNet network

下载: 全尺寸图片幻灯片

图 6 成型网络尾流识别与定位测试结果展示

Figure 6. Test results display of wake identification and location of formed network

下载: 全尺寸图片幻灯片

图 7 两类尾流测试集下不同样本的实时数据记录

Figure 7. Real time data recording of different samples under two types of wake test sets

下载: 全尺寸图片幻灯片

表 1 3种基本网络对比结果

Table 1. Comparison results of three basic networks

Pre-training network	Google	VGG19	AlexNet
Accuracy	83.33%	100%	100%
Time	5min 34s	61min 49s	13min 13s
Stability	Bad	Good	Good

下载: 导出CSV

表 2 Frequency参数对AlexNet网络影响结果

Table 2. Influence results of frequency parameters on AlexNet network

Frequency	1	2	3	4	5
Accuracy	83.33%	100%	100%	100%	100%
Rounds	3	5	5	5	5
Time	8min28s	19min21s	13min13s	12min21s	18min34s
Stability	Better	Better	Better	Good	Better

下载: 导出CSV

表 3 Patience参数对AlexNet网络影响结果

Table 3. Results of influence of patience parameters on AlexNet network

Patience	1	3	5	7
Accuracy	100%	100%	100%	100%
Rounds	2	5	5	5
Time	6min 44s	15min 46s	12min 21s	16min 30s

下载: 导出CSV

表 4 45个样本集/14个测试集（2类）下的实验结果

Table 4. Experimental results under 45 sample sets and 14 test sets (Category 2)

Network type	Accuracy	Missed rate	Error rate	Periscope wake as positive (infrared image)		Ship wake as positive (visible light image)
Network type	Accuracy	Missed rate	Error rate	Precise rate	Recall rate	Precise rate	Recall rate
Untrained	0.5714	35.71%	11.11%	66.67%	100%	100%	85.71%
Transfer learning	0.7857	7.14%	8.97%	100%	75%	81.82%	100%

下载: 导出CSV

表 5 65个样本集/14个测试集下（2类）的实验结果

Table 5. Experimental results of 65 sample sets and 14 test sets (Category 2)

Network type	Accuracy	Missed rate	Error rate	Periscope wake as positive (infrared image)		Ship wake as positive (visible light image)
Network type	Accuracy	Missed rate	Error rate	Precise rate	Recall rate	Precise rate	Recall rate
Untrained	0.7143	7.14%	16.67%	55.56%	100%	100%	87.5%
Transfer learning	0.7857	7.14%	13.63%	71.43%	100%	100%	75%

下载: 导出CSV

表 6 85个样本集/14个测试集下（2类）的实验结果

Table 6. Experimental results of 85 sample sets and 14 test sets (Category 2)

Network type	Accuracy	Missed rate	Error rate	Periscope wake as positive (infrared image)		Ship wake as positive (visible light image)
Network type	Accuracy	Missed rate	Error rate	Precise rate	Recall rate	Precise rate	Recall rate
Untrained	0.3571	14.29%	63.33%	33.33%	75%	83.33%	62.5%
Transfer learning	0.6429	14.29%	14.29%	100%	100%	70%	100%

下载: 导出CSV

参考文献(13)

[1]	王雨农. 基于视觉注意机制的神经网络模型研究及应用[D]. 合肥: 中国科学技术大学, 2017. WANG Yunong. Research on Visual Attention Based Neural Network Model and its Application[D]. Hefei: University of Science and Technology of China, 2017.
[2]	Rumelhart D, Mcclelland J. Learning internal representations by error propagation[M]//Parallel Distributed Processing: Explorations in the Microstructure of Cognition, Massachusetts: MIT Press, 1986: 318-362.
[3]	尹勰, 闫磊. 基于深度卷积神经网络的图像目标检测[J]. 工业控制计算机, 2017, 30(4): 96-97. https://www.cnki.com.cn/Article/CJFDTOTAL-GYKJ201704040.htm YIN Xie, YAN Lei. Image target detection based on deep convolutional neural network [J]. Industrial Control Computer, 2017, 30(4): 96-97. https://www.cnki.com.cn/Article/CJFDTOTAL-GYKJ201704040.htm
[4]	Razavian A S, Azizpour H, Sullivan J, et al. CNN features off-the-shelf: an astounding baseline for recognition[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2014: 512-519.
[5]	Zeiler M D, Fergus R. Visualizing and Understanding Convolutional Networks[M]//Computer Vision-ECCV, Springer International Publishing, 2014.
[6]	胡炎, 单子力, 高峰. 基于Faster-RCNN和多分辨率SAR的海上舰船目标检测[J]. 无线电工程, 2018, 48(2): 96-100. https://www.cnki.com.cn/Article/CJFDTOTAL-WXDG201802005.htm HU Yan, SHAN Zili, GAO Feng. Ship target detection based on faster-RCNN and multi-resolution SAR[J]. Radio Engineering, 2018, 48(2): 96-100. https://www.cnki.com.cn/Article/CJFDTOTAL-WXDG201802005.htm
[7]	李新. 基于红外热像技术连铸板坯裂纹预报方法研究[D]. 唐山: 华北理工大学, 2015. LI Xin. Research on Crack Prediction Method of Continuous Casting Slab Based on Infrared Thermography[D]. Tangshan : North China University of Technology, 2015.
[8]	张健, 杨立, 袁江涛. 水下航行器热尾流试验研究[J]. 实验流体力学, 2008, 22(3): 9-15. https://www.cnki.com.cn/Article/CJFDTOTAL-LTLC200803002.htm ZHANG Jian, YANG Li, YUAN Jiangtao. Experimental study on thermal wake of underwater vehicles[J]. Experimental Fluid Mechanics, 2008, 22(3): 9-15. https://www.cnki.com.cn/Article/CJFDTOTAL-LTLC200803002.htm
[9]	贺林. 水喷淋消声器设计与实验研究[D]. 哈尔滨: 哈尔滨工程大学, 2006. HE Lin. Design and Experimental Study of Water Spray Muffler[D]. Harbin : Harbin Engineering University, 2006.
[10]	伍伟明. 基于Faster R-CNN的目标检测算法的研究[D]. 广州: 华南理工大学, 2018. WU Weiming. Research on Target Detection Algorithm Based on Faster R-CNN[D]. Guangzhou : South China University of Technology, 2018.
[11]	刘万军, 梁雪剑, 曲海成. 自适应增强卷积神经网络图像识别[J]. 中国图象图形学报, 2019, 22(12): 1723-1736. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGTB201712008.htm LIU Wanjun, LIANG Xuejian, QU Haicheng. Adaptive enhanced convolutional neural network image recognition[J]. Chinese Journal of Image Graphics, 2019, 22(12): 1723-1736. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGTB201712008.htm
[12]	Lecun Y, Boser B, Denker J, et al. Back propagation applied to handwritten zip code recognition[J]. Neural Computation, 1989, 1(4): 541-551. doi: 10.1162/neco.1989.1.4.541
[13]	王红霞, 周家奇, 辜承昊, 等. 用于图像分类的卷积神经网络中激活函数的设计[J]. 浙江大学学报: 工学版, 2019, 53(7): 1363-1373. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDZC201907016.htm WANG Hongxia, ZHOU Jiaqi, GU Chenghao, et al. Design of activation functions in convolutional neural networks for image classification[J]. Journal of Zhejiang University: Engineering Edition, 2019, 53(7): 1363-1373. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDZC201907016.htm