Adversarial sample generation method for time-series data based on local augmentation

doi:10.11772/j.issn.1001-9081.2024050610

Journal of Computer Applications ›› 2025, Vol. 45 ›› Issue (5): 1573-1581.DOI: 10.11772/j.issn.1001-9081.2024050610

• Cyber security • Previous Articles

Adversarial sample generation method for time-series data based on local augmentation

Xueying LI¹, Kun YANG², Guoqing TU¹, Shubo LIU²()

^1.Key Laboratory of Aerospace Information Security and Trusted Computing，Ministry of Education，School of Cyber Science and Engineering，Wuhan University，Wuhan Hubei 430072，China
^2.School of Computer Science，Wuhan University，Wuhan Hubei 430072，China

Received:2024-05-14 Revised:2024-07-08 Accepted:2024-08-16 Online:2024-09-04 Published:2025-05-10
Contact: Shubo LIU
About author:LI Xueying， born in 2001， M. S. candidate. Her research interests include time series data， adversarial attack， machine learning， artificial intelligence security.
YANG Kun， born in 1992， M. S. candidate. His research interests include embedded system， machine learning， cryptography， internet of things security， artificial intelligence security.
TU Guoqing， born in 1974， Ph. D.， associate professor. His research interests include embedded system， internet of things， information security， water conservancy informatization.
LIU Shubo， born in 1970， Ph. D.， professor. His research interests include embedded system， differential privacy， time series.
Supported by:
National Natural Science Foundation of China(41971407)

基于局部增强的时序数据对抗样本生成方法

李雪莹¹, 杨琨², 涂国庆¹, 刘树波²()

^1.空天信息安全与可信计算教育部重点实验室，武汉大学国家网络安全学院，武汉 430072
^2.武汉大学计算机学院，武汉 430072

通讯作者: 刘树波
作者简介:李雪莹（2001—），女，山东莘县人，硕士研究生，主要研究方向：时序数据、对抗攻击、机器学习、人工智能安全
杨琨（1992—），男，安徽安庆人，硕士研究生，主要研究方向：嵌入式系统、机器学习、密码学、物联网安全、人工智能安全
涂国庆（1974—），男，湖北罗田人，副教授，博士，主要研究方向：嵌入式系统、物联网、信息安全、水利信息化
刘树波（1970—），男，黑龙江泰来人，教授，博士，主要研究方向：嵌入式系统、差分隐私、时间序列。
基金资助:
国家自然科学基金资助项目(41971407)

Abstract

Abstract:

Deep Neural Networks （DNNs） are highly susceptible to adversarial attacks， causing security problems in time-series data classification tasks. Gradient-based attack methods can generate adversarial samples quickly but need continuous access to the model's internal information， while generation-based attack methods do not need this access after training but suffer from poor stealthiness and transferability. To address these problems， a semi-white box adversarial sample generation method for time-series data based on local augmentation was proposed using the generative attack method AdvGAN. The local augmentation strategy in this method injected information from other data categories into original samples and utilized enhanced data to execute semi-white-box attacks. The attack model leveraged both original sample information and distribution characteristics of other categories， thereby enhancing model's attack capability and transferability. Experimental results on UCR datasets demonstrate that the proposed method generates an adversarial example in 0.027 s on average； it outperforms Fast Gradient Sign Method （FGSM）， AdvGAN， and GATN （Gradient Adversarial Transformation Network） methods in attack success rate on 18， 25， and 13 datasets of 27 datasets respectively. The generated adversarial examples exhibit significantly smaller Mean Squared Error （MSE） compared to AdvGAN and GATN methods on 20 and 27 datasets respectively. Its transfer success rates surpass AdvGAN and FGSM methods on 18 and 11 datasets respectively， with transfer attack success rates exceeding 25% on 9 datasets of 21 datasets. The results indicate that the proposed method maintains efficient adversarial example generation while improving stealthiness and preserving competitive attack performance.

Key words: time-series data, deep learning, adversarial sample, data augmentation, semi-white box attack

摘要：

深度神经网络（DNN）极易遭受对抗攻击，进而引起时序数据分类任务中的安全问题。基于梯度的攻击方法可以快速地生成对抗样本，但需要不断访问模型内部信息；基于生成的攻击方法在模型训练完成之后无须访问模型内部信息，但存在隐蔽性和迁移性较差等问题。针对以上问题，基于生成式攻击方法AdvGAN提出一种基于局部增强的时序数据对抗样本生成方法，其中的局部增强策略将其他类别数据的信息注入原样本中，并利用增强后的数据执行灰盒攻击；而攻击模型不仅可以利用原样本信息，还能利用其他类别样本的分布信息，进而提升模型的攻击能力和迁移能力。在UCR数据集上的实验结果表明，所提方法平均0.027 s即可生成一个对抗样本；在27个数据集中，它的攻击成功率分别在18、25和13个数据集上优于快速梯度符号法（FGSM）、AdvGAN和GATN（Gradient Adversarial Transformation Network）方法；它的生成对抗样本的均方误差（MSE）分别在20和27个数据集上明显小于AdvGAN和GATN方法；在21个数据集中，它的迁移成功率分别在18和11个数据集上优于AdvGAN和FGSM方法，且在9个数据集上的迁移攻击成功率达到25%以上。可见，所提方法在保证对抗样本生成速度的同时，提高了对抗样本的隐蔽性并保持有竞争力的攻击效果。

关键词: 时序数据, 深度学习, 对抗样本, 数据增强, 灰盒攻击

CLC Number:

TP183

Xueying LI, Kun YANG, Guoqing TU, Shubo LIU. Adversarial sample generation method for time-series data based on local augmentation[J]. Journal of Computer Applications, 2025, 45(5): 1573-1581.

李雪莹, 杨琨, 涂国庆, 刘树波. 基于局部增强的时序数据对抗样本生成方法[J]. 《计算机应用》唯一官方网站, 2025, 45(5): 1573-1581.

Figures/Tables 10

Fig. 1 Model structure of proposed method

Fig. 2 Structure of CVAE network

Fig. 3 Structure of generator network

Fig. 4 Structure of discriminator network

Tab. 1 Average time for adversarial sample generation on different datasets

数据集	数据长度	AdvGAN	本文方法	FGSM
Car	577	0.011 0	0.040 0	0.045
ECG5000	140	0.032 0	0.028 0	0.067
FordB	500	0.008 6	0.010 0	0.045
Lightning2	637	0.011 0	0.044 0	0.056
Lightning7	319	0.006 8	0.012 5	0.020
Plane	144	0.004 2	0.009 0	0.080
MoteStrain	84	0.011 0	0.055 0	0.120
Trace	275	0.006 4	0.014 2	0.080

Fig. 5 Comparison of IMSE of adversarial samples on 27 datasets

Tab. 2 Comparison of success rate of attacking FCN model on Dtest dataset

数据类型	数据集	原始分类准确率/%	$I A S R$
数据类型	数据集	原始分类准确率/%	FGSM	本文方法	AdvGAN	GATN
Sensor	Car	57.00	58.82	58.82	58.82	96
	Chlorine	81.65	98.00	75.20	71.63	87
	CinCETorso	86.33	60.80	58.79	41.47	88
	Earthquakes	74.00	54.90	29.41	17.64	31
ECG	ECG200	91.00	25.27	41.75	40.66	23
	ECG5000	93.89	17.07	27.10	22.73	93
	ECGFiveDays	99.00	47.06	50.30	49.25	18
Sensor	FordA	93.14	48.35	48.35	48.35	48
	FordB	79.19	46.54	46.54	46.54	55
	Insect	37.80	85.95	87.46	85.62	75
	Italy	95.90	5.06	9.30	7.40	2
	Lightning2	71.00	54.50	22.73	27.27	53
	Lightning7	83.33	76.67	80.00	66.67	67
	StarLightCurves	96.61	15.47	10.10	4.03	22
	MoteStrain	92.52	8.30	10.35	8.60	5
	AllGestureX	54.69	61.56	69.03	45.52	25
	AllGestureY	73.67	83.10	78.67	62.33	53
	AllGestureZ	70.61	76.01	77.17	63.87	43
	Phoneme	32.31	50.31	48.41	41.40	76
	Plane	96.88	13.20	81.13	84.90	72
	Sony1	72.92	14.57	27.14	25.71	46
	Sony2	91.34	6.90	17.67	16.38	4
	Trace	100.00	18.00	68.00	70.00	100
	Wafer	99.16	89.50	88.66	88.33	97
	FreezerRTrain	90.66	44.80	44.85	44.85	0
	FreezerSTrain	71.79	24.39	24.12	21.95	52
	DLWeekend	31.88	—	4.55	—	78

Tab. 2 Comparison of success rate of attacking FCN model on Dtest dataset

数据类型	数据集	原始分类准确率/%	$I A S R$
数据类型	数据集	原始分类准确率/%	FGSM	本文方法	AdvGAN	GATN
Sensor	Car	57.00	58.82	58.82	58.82	96
	Chlorine	81.65	98.00	75.20	71.63	87
	CinCETorso	86.33	60.80	58.79	41.47	88
	Earthquakes	74.00	54.90	29.41	17.64	31
ECG	ECG200	91.00	25.27	41.75	40.66	23
	ECG5000	93.89	17.07	27.10	22.73	93
	ECGFiveDays	99.00	47.06	50.30	49.25	18
Sensor	FordA	93.14	48.35	48.35	48.35	48
	FordB	79.19	46.54	46.54	46.54	55
	Insect	37.80	85.95	87.46	85.62	75
	Italy	95.90	5.06	9.30	7.40	2
	Lightning2	71.00	54.50	22.73	27.27	53
	Lightning7	83.33	76.67	80.00	66.67	67
	StarLightCurves	96.61	15.47	10.10	4.03	22
	MoteStrain	92.52	8.30	10.35	8.60	5
	AllGestureX	54.69	61.56	69.03	45.52	25
	AllGestureY	73.67	83.10	78.67	62.33	53
	AllGestureZ	70.61	76.01	77.17	63.87	43
	Phoneme	32.31	50.31	48.41	41.40	76
	Plane	96.88	13.20	81.13	84.90	72
	Sony1	72.92	14.57	27.14	25.71	46
	Sony2	91.34	6.90	17.67	16.38	4
	Trace	100.00	18.00	68.00	70.00	100
	Wafer	99.16	89.50	88.66	88.33	97
	FreezerRTrain	90.66	44.80	44.85	44.85	0
	FreezerSTrain	71.79	24.39	24.12	21.95	52
	DLWeekend	31.88	—	4.55	—	78

Tab. 3 Performance comparison of transfer attack against ResNet model on Dtest dataset

数据集	原始分类准确率/%	本文方法		AdvGAN		FGSM
数据集	原始分类准确率/%	$I A S R$ /%	$I M S E$	$I A S R$ /%	$I M S E$	$I A S R$ /%	$I M S E$
Earthquakes	75.36	5.77	0.003 5	1.92	0.003 7	3.80	0.002 5
ECG200	88.00	1.14	0.003 3	2.27	0.001 0	6.80	0.002 5
ECG5000	94.36	17.88	0.005 5	5.28	0.001 8	5.90	0.002 5
ECGFiveDays	94.20	39.26	0.007 7	37.15	0.027 9	13.03	0.002 5
FordB	81.13	6.74	0.007 0	1.09	0.001 0	4.35	0.002 5
Insect	45.10	78.04	0.006 3	73.71	0.033 5	69.65	0.002 5
Italy	95.20	4.50	0.010 0	3.67	0.027 0	2.45	0.002 5
Lightning2	64.52	10.00	0.005 0	10.00	0.007 0	25.00	0.002 5
Lightning7	80.56	68.97	0.031 0	44.83	0.045 0	48.00	0.002 5
StarLightCurves	97.12	12.56	0.000 3	6.60	0.000 9	15.19	0.002 5
MoteStrain	92.32	6.40	0.010 0	3.03	0.004 9	5.40	0.002 5
AllGestureX	63.47	52.73	0.000 7	35.04	0.001 6	61.41	0.002 5
AllGestureY	71.22	76.50	0.000 8	26.65	0.000 2	75.35	0.002 5
AllGestureZ	69.00	46.75	0.001 3	38.46	0.001 7	58.28	0.002 5
Phoneme	32.10	25.63	0.010 0	28.21	0.012 0	35.26	0.002 5
Sony1	96.88	1.10	0.002 8	0.86	0.006 8	4.50	0.002 5
Sony2	91.60	8.02	0.007 0	11.30	0.021 0	3.90	0.002 5
Trace	100.00	26.00	0.001 9	10.00	0.001 8	37.00	0.002 5
Wafer	99.33	1.69	0.000 6	0.67	0.000 8	16.05	0.002 5
FreezerRTrain	98.64	48.71	0.001 2	48.52	0.022 4	48.71	0.002 5
FreezerSTrain	72.57	21.98	0.005 9	13.94	0.004 9	26.27	0.002 5

Tab. 3 Performance comparison of transfer attack against ResNet model on Dtest dataset

数据集	原始分类准确率/%	本文方法		AdvGAN		FGSM
数据集	原始分类准确率/%	$I A S R$ /%	$I M S E$	$I A S R$ /%	$I M S E$	$I A S R$ /%	$I M S E$
Earthquakes	75.36	5.77	0.003 5	1.92	0.003 7	3.80	0.002 5
ECG200	88.00	1.14	0.003 3	2.27	0.001 0	6.80	0.002 5
ECG5000	94.36	17.88	0.005 5	5.28	0.001 8	5.90	0.002 5
ECGFiveDays	94.20	39.26	0.007 7	37.15	0.027 9	13.03	0.002 5
FordB	81.13	6.74	0.007 0	1.09	0.001 0	4.35	0.002 5
Insect	45.10	78.04	0.006 3	73.71	0.033 5	69.65	0.002 5
Italy	95.20	4.50	0.010 0	3.67	0.027 0	2.45	0.002 5
Lightning2	64.52	10.00	0.005 0	10.00	0.007 0	25.00	0.002 5
Lightning7	80.56	68.97	0.031 0	44.83	0.045 0	48.00	0.002 5
StarLightCurves	97.12	12.56	0.000 3	6.60	0.000 9	15.19	0.002 5
MoteStrain	92.32	6.40	0.010 0	3.03	0.004 9	5.40	0.002 5
AllGestureX	63.47	52.73	0.000 7	35.04	0.001 6	61.41	0.002 5
AllGestureY	71.22	76.50	0.000 8	26.65	0.000 2	75.35	0.002 5
AllGestureZ	69.00	46.75	0.001 3	38.46	0.001 7	58.28	0.002 5
Phoneme	32.10	25.63	0.010 0	28.21	0.012 0	35.26	0.002 5
Sony1	96.88	1.10	0.002 8	0.86	0.006 8	4.50	0.002 5
Sony2	91.60	8.02	0.007 0	11.30	0.021 0	3.90	0.002 5
Trace	100.00	26.00	0.001 9	10.00	0.001 8	37.00	0.002 5
Wafer	99.33	1.69	0.000 6	0.67	0.000 8	16.05	0.002 5
FreezerRTrain	98.64	48.71	0.001 2	48.52	0.022 4	48.71	0.002 5
FreezerSTrain	72.57	21.98	0.005 9	13.94	0.004 9	26.27	0.002 5

Fig. 6 Comparison of attack success rate on different datasets before and after augmentation on FCN

Fig. 7 Comparison of success rate of transfer attack before and after augmentation on ResNet

References 36

1	任利强，贾舒宜，王海鹏，等. 基于深度学习的时间序列分类研究综述［J］. 电子与信息学报， 2024， 46（8）： 3094-3116.
	REN L Q， JIA S Y， WANG H P， et al. A review of deep learning-based time series classification research［J］. Journal of Electronics and Information Technology， 2024， 46（8）： 3094-3116.
2	SUN L， ZHANG M， WANG B， et al. Few-shot class-incremental learning for medical time series classification［J］. IEEE Journal of Biomedical and Health Informatics， 2024， 28（4）： 1872-1882.
3	LUCAS B， SHIFAZ A， PELLETIER C， et al. Proximity Forest： an effective and scalable distance-based classifier for time series［J］. Data Mining and Knowledge Discovery， 2019， 33（3）：607-635.
4	宋钰丹，王晶，王雪徽，等. 基于自适应多任务学习的睡眠生理时序分类方法［J］. 计算机应用， 2024， 44（2）： 654-662.
	SONG Y D， WANG J， WANG X H， et al. Sleep physiological time series classification method based on adaptive multi-task learning［J］. Journal of Computer Applications， 2024， 44（2）： 654-662.
5	张伟雄，唐娉，张正. 基于时序自注意力机制的遥感数据时间序列分类［J］. 遥感学报， 2023， 27（8）： 1914-1924.
	ZHANG W X， TANG P， ZHANG Z. Time series classification of remote sensing data based on temporal self-attention mechanism［J］. National Remote Sensing Bulletin， 2023， 27（8）： 1914-1924.
6	SZEGEDY C， ZAREMBA W， SUTSKEVER I， et al. Intriguing properties of neural networks［EB/OL］. ［2024-01-20］..
7	刘西蒙，谢乐辉，王耀鹏，等. 深度学习中的对抗攻击与防御［J］. 网络与信息安全学报， 2020， 6（5）： 36-53.
	LIU X M， XIE L H， WANG Y P， et al. Adversarial attacks and defenses in deep learning［J］. Chinese Journal of Network and Information Security， 2020， 6（5）： 36-53.
8	GOOFELLOW J I， SHLENS J， SZEGEDY C. Explaining and harnessing adversarial examples［EB/OL］. ［2023-12-18］..
9	XIAO C， LI B， ZHU J Y， et al. Generating adversarial examples with adversarial networks［C］// Proceedings of the 27th International Joint Conference on Artificial Intelligence. San Francisco： Morgan Kaufmann Publishers Inc.， 2018： 3905-3911.
10	YANG W， YUAN J， WANG X， et al. TSadv： black-box adversarial attack on time series with local perturbations［J］. Engineering Applications of Artificial Intelligence， 2022， 114： No.105218.
11	OREGI I， DEL SER J， PEREZ A， et al. Adversarial sample crafting for time series classification with elastic similarity measures［C］// Intelligent distributed computing Ⅻ： 12th International Symposium on Intelligent Distributed Computing， SCI 798. Cham： Springer， 2018：26-39.
12	FAWAZ H I， FORESTIER G， WEBER J， et al. Adversarial attacks on deep neural networks for time series classification［C］// Proceedings of the 2019 International Joint Conference on Neural Networks. Piscataway： IEEE， 2019：1-8.
13	KURAKIN A， GOODFELLOW I， BENGIO S. Adversarial machine learning at scale［EB/OL］. ［2024-02-01］..
14	PIALLA G， FAWAZ H I， DEVANNE M， et al. Smooth perturbations for time series adversarial attacks［C］// Proceedings of the 2022 Pacific-Asia Conference on Knowledge Discovery and Data Mining， LNCS 13280. Cham： Springer， 2022：458-496.
15	DONG C G， ZHENG L N， CHEN W， et al. SWAP： exploiting second-ranked logits for adversarial attacks on time series［C］// Proceedings of the 2023 IEEE International Conference on Knowledge Graph. Piscataway： IEEE， 2023：117-125.
16	WU T， WANG X， QIAO S， et al. Small perturbations are enough： adversarial attacks on time series prediction［J］. Information Sciences， 2022， 587： 794-812.
17	CHEN Z， DOST K， ZHU X， et al. Targeted attacks on time series forecasting［C］// Proceedings of the 2023 Pacific-Asia Conference on Knowledge Discovery and Data Mining， LNCS 13938. Cham： Springer， 2023：314-327.
18	KARIM F， MAJUMDAR S， DARABI H. Adversarial attacks on time series［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2021， 43（10）：3309-3320.
19	HARFORD S， KARIM F， DARABI H. Generating adversarial samples on multivariate time series using variational autoencoders［J］. IEEE/CAA Journal of Automatica Sinica， 2021， 8（9）：1523-1538.
20	ZHANG H， CISSE M， DAUPHIN Y N， et al. mixup： beyond empirical risk minimization［EB/OL］. ［2024-01-11］..
21	KINGMA P D， WELLING M. Auto-encoding variational Bayes［EB/OL］. ［2024-03-17］..
22	曹春明，何戡，宗学军，等. 基于VAE和DLIESN的工控系统入侵检测方法［J］. 计算机工程与设计， 2023， 44（11）： 3283-3289.
	CAO C M， HE K， ZONG X J， et al. Intrusion detection method of industrial control system based on VAE and DLIESN［J］. Computer Engineering and Design， 2023， 44（11）： 3283-3289.
23	范振杰，罗娜. 基于改进VAE的时间序列数据增强方法［J］. 华东理工大学学报（自然科学版）， 2024， 50（3）： 400-410.
	FAN Z J， LUO N. A time series data enhancement method based on improved VAE［J］. Journal of East China University of Science and Technology， 2024， 50（3）： 400-410.
24	张子宣，宗学军，何戡，等. 基于CVAE-CatBoost的工业控制网络异常流量检测研究［J］. 计算机工程， 2023， 49（5）： 173-180.
	ZHANG Z X， ZONG X J， HE K， et al. Research on abnormal traffic detection in industrial control network based on CVAE-CatBoost［J］. Computer Engineering， 2023， 49（5）： 173-180.
25	FENG W， XU N， ZHANG T， et al. Dynamic generative targeted attacks with pattern injection［C］// Proceedings of 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2023：16404-16414.
26	WANG Z， YANG H， FENG Y， et al. Towards transferable targeted adversarial examples［C］// Proceedings of 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2023：20534-20543.
27	CARLINI N， WAGNER D. Towards evaluating the robustness of neural networks［C］// Proceedings of 2017 IEEE Symposium on Security and Privacy. Piscataway： IEEE， 2017：39-57.
28	HU J， SHEN L， SUN G. Squeeze-and-excitation networks［C］// Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2018：7132-7141.
29	HE K， GKIOXARI G， DOLLÁR P， et al. Mask R-CNN［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2020， 42（2）：386-397.
30	ZHOU B， KHOSLA A， LAPEDRIZA À， et al. Learning deep features for discriminative localization［C］// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2018： 2921-2929.
31	YE L， KEOGH E. Time series shapelets： a novel technique that allows accurate， interpretable and fast classification［J］. Data Mining and Knowledge Discovery， 2011， 22（1/2）：149-182.
32	CHUNG F L， FU T C， NG V， et al. An evolutionary approach to pattern-based time series segmentation［J］. IEEE Transactions on Evolutionary Computation， 2004， 8（5）：471-489.
33	BREIMAN L. Random forests［J］. Machine Learning， 2001， 45（1）： 5-32.
34	DAU H A， BAGNALL A， KAMGAR K， et al. The UCR time series archive［J］. IEEE/CAA Journal of Automatica Sinica， 2019， 6（6）：1293-1305.
35	WANG Z， YAN W， OATES T. Time series classification from scratch with deep neural networks： a strong baseline［C］// Proceedings of the 2017 International Joint Conference on Neural Networks. Piscataway： IEEE， 2017：1578-1585.
36	HE K， ZHANG X， REN S， et al. Deep residual learning for image recognition［C］// Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2016： 770-778.

[1]	Sijie NIU, Yuliang LIU. Auxiliary diagnostic method for retinopathy based on dual-branch structure with knowledge distillation [J]. Journal of Computer Applications, 2025, 45(5): 1410-1414.
[2]	Lihu PAN, Shouxin PENG, Rui ZHANG, Zhiyang XUE, Xuzhen MAO. Video anomaly detection for moving foreground regions [J]. Journal of Computer Applications, 2025, 45(4): 1300-1309.
[3]	Yiding WANG, Zehao WANG, Yaoli LI, Shaoqing CAI, Yuan YUAN. Multi-scale 2D-Adaboost microscopic image recognition algorithm of Chinese medicinal materials powder [J]. Journal of Computer Applications, 2025, 45(4): 1325-1332.
[4]	Renjie TIAN, Mingli JING, Long JIAO, Fei WANG. Recommendation algorithm of graph contrastive learning based on hybrid negative sampling [J]. Journal of Computer Applications, 2025, 45(4): 1053-1060.
[5]	Haitao SUN, Jiayu LIN, Zuhong LIANG, Jie GUO. Data augmentation technique incorporating label confusion for Chinese text classification [J]. Journal of Computer Applications, 2025, 45(4): 1113-1119.
[6]	Yang ZHOU, Hui LI. Remote sensing image building extraction network based on dual promotion of semantic and detailed features [J]. Journal of Computer Applications, 2025, 45(4): 1310-1316.
[7]	Zhenhua XUE, Qiang LI, Chao HUANG. Vision foundation model-driven pixel-level image anomaly detection method [J]. Journal of Computer Applications, 2025, 45(3): 823-831.
[8]	Ruilong CHEN, Tao HU, Youjun BU, Peng YI, Xianjun HU, Wei QIAO. Stacking ensemble adversarial defense method for encrypted malicious traffic detection model [J]. Journal of Computer Applications, 2025, 45(3): 864-871.
[9]	Chenwei SUN, Junli HOU, Xianggen LIU, Jiancheng LYU. Large language model prompt generation method for engineering drawing understanding [J]. Journal of Computer Applications, 2025, 45(3): 801-807.
[10]	Miaolei DENG, Yupei KAN, Chuanchuan SUN, Haihang XU, Shaojun FAN, Xin ZHOU. Summary of network intrusion detection systems based on deep learning [J]. Journal of Computer Applications, 2025, 45(2): 453-466.
[11]	Songsen YU, Zhifan LIN, Guopeng XUE, Jianyu XU. Lightweight large-format tile defect detection algorithm based on improved YOLOv8 [J]. Journal of Computer Applications, 2025, 45(2): 647-654.
[12]	Danni DING, Bo PENG, Xi WU. VPNet： fatty liver ultrasound image classification method inspired by ventral pathway [J]. Journal of Computer Applications, 2025, 45(2): 662-669.
[13]	Zirong HONG, Guangqing BAO. Review of radar automatic target recognition based on ensemble learning [J]. Journal of Computer Applications, 2025, 45(2): 371-382.
[14]	Zhongwei ZHANG, Jun WANG, Shudong LIU, Zhiheng WANG. Object detection in remote sensing image based on multi-scale feature fusion and weighted boxes fusion [J]. Journal of Computer Applications, 2025, 45(2): 633-639.
[15]	Xuewen YAN, Zhangjin HUANG. Few-shot image classification method based on contrast learning [J]. Journal of Computer Applications, 2025, 45(2): 383-391.

Adversarial sample generation method for time-series data based on local augmentation

基于局部增强的时序数据对抗样本生成方法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 36

Related Articles 15

Recommended Articles

Metrics