基于交叉细化和循环注意力的RGB-D显著性目标检测

doi:10.16182/j.issn1004731x.joss.22-1372

摘要/Abstract

摘要：

针对显著性目标检测区域边界模糊以及检测区域不精确不完整的问题，提出了基于交叉细化和循环注意力的RGB-D显著性目标检测方法。在利用编码器提取特征的阶段设计了交叉细化模块，用于补充对方的特征信息，改善了融合前的特征质量，抑制了质量较差的深度图带来的消极影响，解决了显著性目标边缘模糊的问题。针对融合后的特征，提出联合注意力机制与卷积长短期记忆网络单元的循环模块以模拟大脑的内部生成机制，通过检索过往的记忆帮助推断当前的决策，从而获得需要长期记忆的语义场景，可以全面学习融合特征的内部语义关系，生成检测区域更完整，更准确的显著性图。在6个公开数据集上进行的实验表明，所提的方法可以得到边缘清晰且准确度更高的显著图。

关键词: RGB-D, 显著性目标检测, 交叉细化, 注意力机制, 卷积长短期记忆网络, 循环模块

Abstract:

In order to solve the problems that the boundary of the saliency object detection area is vague, and the detection area is incomplete or inaccurate, an RGB-D saliency object detection method based on cross-refinement and circular attention is proposed. A cross-refinement module is designed at the stage of extracting features using encoders, which is used to supplement feature information of each other and improve the feature quality before fusion. It also suppresses the negative impact of poor-quality depth maps and addresses the issue that the edges of the saliency object are blurred. For the features after fusion, the circular module is proposed, which combines the attention mechanism with convolutional long short-term memory (LSTM) network unit to simulate the internal generation mechanism of the brain and help infer the current decision by retrieving past memories, so as to obtain semantic scenes that require long-term memory. The module can comprehensively learn the internal semantic relationships of fusion features to generate a more complete and accurate saliency map for the detection area. Experiments conducted on six public datasets show that the proposed method can obtain a saliency map with clear edges and high accuracy.

Key words: RGB-D, saliency object detection, cross-refinement, attention mechanism, convolutional long short-term memory network, circular module

中图分类号:

TP391.9

董庆庆,吴昊,钱文华等 . 基于交叉细化和循环注意力的RGB-D显著性目标检测[J]. 系统仿真学报, 2023, 35(9): 1931-1947.

Dong Qingqing,Wu Hao,Qian Wenhua,et al . RGB-D Saliency Object Detection Based on Cross-refinement and Circular Attention[J]. Journal of System Simulation, 2023, 35(9): 1931-1947.

图/表 16

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

表1

本文方法与11种模型在6种指标下的定量比较

数据集	指标	JLDCF	HAINet	CDNet	DSA^2F	JSM	Mobile Sal	DSU	CFIDnet	DCMF	SSL	C²DFnet	Ours
DES	$M A E ↓$	0.021	0.018	0.020	0.023	0.058	0.021	0.063	0.023	0.023	0.019	0.020	0.016
	$F β m a x ↑$	0.923	0.936	0.936	0.916	0.808	0.924	0.797	0.911	0.924	0.936	0.916	0.937
	$F β a v g ↑$	0.907	0.921	0.899	0.904	0.762	0.908	0.753	0.898	0.899	0.919	0.907	0.922
	$F β ω ↑$	0.889	0.910	0.863	0.889	0.742	0.896	0.682	0.886	0.879	0.908	0.898	0.910
	$s α ↑$	0.931	0.931	0.931	0.917	0.820	0.929	0.818	0.917	0.932	0.932	0.922	0.933
	$E ξ ↑$	0.968	0.971	0.971	0.954	0.875	0.970	0.888	0.940	0.968	0.972	0.959	0.973
LFSD	$M A E ↓$	0.070	0.079	0.088	0.055	0.127	0.080	0.129	0.071	0.069	0.079	0.065	0.070
	$F β m a x ↑$	0.867	0.853	0.841	0.889	0.781	0.841	0.792	0.865	0.875	0.835	0.867	0.869
	$F β a v g ↑$	0.848	0.842	0.822	0.879	0.757	0.825	0.759	0.850	0.847	0.826	0.859	0.859
	$F β ω ↑$	0.805	0.803	0.775	0.848	0.724	0.780	0.710	0.807	0.806	0.791	0.829	0.826
	$s α ↑$	0.861	0.854	0.842	0.883	0.767	0.847	0.778	0.869	0.877	0.844	0.864	0.878
	$E ξ ↑$	0.902	0.886	0.877	0.924	0.833	0.888	0.842	0.903	0.909	0.879	0.903	0.913
NJU2K	$M A E ↓$	0.036	0.038	0.060	0.040	0.133	0.047	0.135	0.038	0.051	0.043	0.035	0.029
	$F β m a x ↑$	0.915	0.915	0.871	0.907	0.710	0.896	0.720	0.915	0.873	0.902	0.919	0.926
	$F β a v g ↑$	0.902	0.903	0.847	0.898	0.673	0.875	0.683	0.899	0.868	0.887	0.897	0.916
	$F β ω ↑$	0.880	0.878	0.804	0.878	0.647	0.840	0.631	0.871	0.844	0.861	0.878	0.898
	$s α ↑$	0.912	0.912	0.876	0.904	0.713	0.895	0.727	0.914	0.871	0.901	0.908	0.925
	$E ξ ↑$	0.950	0.944	0.911	0.938	0.801	0.932	0.805	0.946	0.915	0.938	0.943	0.956
SIP	$M A E ↓$	0.049	0.053	0.059	0.057	0.146	0.053	0.156	0.060	0.043	0.058	0.053	0.043
	$F β m a x ↑$	0.889	0.892	0.878	0.875	0.645	0.880	0.624	0.870	0.899	0.874	0.877	0.902
	$F β a v g ↑$	0.873	0.876	0.848	0.866	0.622	0.862	0.605	0.857	0.891	0.862	0.864	0.889
	$F β ω ↑$	0.845	0.845	0.805	0.840	0.562	0.834	0.520	0.829	0.872	0.841	0.834	0.866
	$s α ↑$	0.880	0.880	0.870	0.862	0.697	0.873	0.687	0.864	0.886	0.868	0.872	0.892
	$E ξ ↑$	0.924	0.922	0.914	0.912	0.774	0.916	0.766	0.909	0.930	0.910	0.916	0.931
NLPR^]	$M A E ↓$	0.023	0.024	0.062	0.024	0.060	0.025	0.065	0.026	0.029	0.027	0.022	0.021
	$F β m a x ↑$	0.917	0.915	0.810	0.906	0.771	0.908	0.765	0.905	0.906	0.899	0.918	0.924
	$F β a v g ↑$	0.894	0.901	0.757	0.896	0.743	0.886	0.734	0.892	0.872	0.882	0.903	0.910
	$F β ω ↑$	0.866	0.877	0.684	0.873	0.706	0.858	0.652	0.867	0.836	0.859	0.882	0.890
	$s α ↑$	0.925	0.924	0.850	0.919	0.805	0.920	0.807	0.922	0.922	0.913	0.928	0.931
	$E ξ ↑$	0.963	0.960	0.892	0.952	0.875	0.961	0.875	0.955	0.954	0.949	0.963	0.965
STERE	$M A E ↓$	0.040	0.040	0.059	0.039	0.096	0.041	0.099	0.043	0.043	0.047	0.039	0.037
	$F β m a x ↑$	0.904	0.906	0.870	0.900	0.772	0.895	0.779	0.897	0.906	0.883	0.897	0.907
	$F β a v g ↑$	0.874	0.887	0.831	0.889	0.747	0.875	0.752	0.887	0.869	0.867	0.881	0.891
	$F β ω ↑$	0.831	0.852	0.769	0.866	0.709	0.842	0.689	0.850	0.825	0.837	0.849	0.860
	$s α ↑$	0.903	0.907	0.871	0.898	0.783	0.903	0.789	0.901	0.910	0.885	0.902	0.908
	$E ξ ↑$	0.945	0.944	0.922	0.942	0.852	0.940	0.861	0.942	0.946	0.930	0.943	0.947

表1

图11

表2

对本文方法的定量消融实验结果

模型	SIP				NLPR				STERE
模型	$M A E ↓$	$F β m a x ↑$	$s α ↑$	$E ξ ↑$	$M A E ↓$	$F β m a x ↑$	$s α ↑$	$E ξ ↑$	$M A E ↓$	$F β m a x ↑$	$s α ↑$	$E ξ ↑$
Baseline	0.044	0.889	0.878	0.918	0.023	0.915	0.922	0.960	0.039	0.893	0.896	0.938
+MCR	0.043	0.899	0.886	0.927	0.022	0.919	0.928	0.962	0.038	0.900	0.903	0.942
+RCL	0.043	0.900	0.889	0.929	0.022	0.921	0.927	0.963	0.038	0.902	0.904	0.940
Ours	0.043	0.902	0.892	0.931	0.021	0.924	0.931	0.965	0.037	0.907	0.908	0.947

表2

表3

MCR模块中多模态交互细化策略的消融分析

不同策略的MCR	SIP				NLPR				STERE
不同策略的MCR	$M A E ↓$	$F β m a x ↑$	$s α ↑$	$E ξ ↑$	$M A E ↓$	$F β m a x ↑$	$s α ↑$	$E ξ ↑$	$M A E ↓$	$F β m a x ↑$	$s α ↑$	$E ξ ↑$
Cat	0.044	0.898	0.891	0.931	0.022	0.924	0.929	0.963	0.037	0.906	0.908	0.946
Mul	0.044	0.899	0.890	0.930	0.022	0.923	0.927	0.961	0.038	0.905	0.906	0.943
Mul+Cat	0.045	0.895	0.889	0.929	0.022	0.922	0.926	0.962	0.038	0.904	0.905	0.942
Ours	0.043	0.902	0.892	0.931	0.021	0.924	0.931	0.965	0.037	0.907	0.908	0.947

表3

表4

对主干网络的有效性分析

骨干网络	SIP				NLPR				STERE
骨干网络	$M A E ↓$	$F β m a x ↑$	$s α ↑$	$E ξ ↑$	$M A E ↓$	$F β m a x ↑$	$s α ↑$	$E ξ ↑$	$M A E ↓$	$F β m a x ↑$	$s α ↑$	$E ξ ↑$
VGG11	0.047	0.879	0.869	0.882	0.027	0.909	0.911	0.955	0.043	0.873	0.876	0.918
VGG16	0.045	0.882	0.872	0.898	0.025	0.915	0.918	0.957	0.041	0.881	0.882	0.922
ResNet34	0.044	0.899	0.890	0.925	0.024	0.922	0.929	0.960	0.039	0.899	0.900	0.939
ResNet50	0.043	0.902	0.892	0.931	0.021	0.924	0.931	0.965	0.037	0.907	0.908	0.947

表4

表5

本文方法与原始模型的定量比较

模型	HAINet				C²DFNet^[46]
模型	$M A E ↓$	$F β m a x ↑$	$s α ↑$	$E ξ ↑$	$M A E ↓$	$F β m a x ↑$	$s α ↑$	$E ξ ↑$
Baseline	0.038	0.915	0.912	0.944	0.035	0.919	0.908	0.943
+MCR	0.034	0.920	0.919	0.948	0.031	0.923	0.917	0.953
+RCL	0.033	0.921	0.922	0.948	0.033	0.921	0.915	0.950

表5

参考文献 47

1	Ren Zhixiang, Gao Shenghua, Chia L T, et al. Region-based Saliency Detection and Its Application in Object Recognition[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2014, 24(5): 769-779.
2	Wei Yunchao, Liang Xiaodan, Chen Yunpeng, et al. STC: A Simple to Complex Framework for Weakly-supervised Semantic Segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(11): 2314-2320.
3	Fan Dengping, Cheng Mingming, Liu Yun, et al. Structure-measure: A New Way to Evaluate Foreground Maps[C]//2017 IEEE International Conference on Computer Vision (ICCV). Piscataway, NJ, USA: IEEE, 2017: 4558-4567.
4	Mahadevan V, Vasconcelos N. Saliency-based Discriminant Tracking[C]//2009 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2009: 1007-1013.
5	Gao Dashan, Han S, Vasconcelos N. Discriminant Saliency, the Detection of Suspicious Coincidences, and Applications to Visual Recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 31(6): 989-1005.
6	Ciptadi A, Hermans T, Rehg J M. An in Depth View of Saliency[C]//Proceedings of the British Machine Vision Conference. Durham, UK: BMVA Press, 2013: 112.1-112.11.
7	Cheng Yupeng, Fu Huazhu, Wei Xingxing, et al. Depth Enhanced Saliency Detection Method[C]//Proceedings of International Conference on Internet Multimedia Computing and Service. New York, NY, USA: Association for Computing Machinery, 2014: 23-27.
8	Ren Jianqiang, Gong Xiaojin, Yu Lu, et al. Exploiting Global Priors for RGB-D Saliency Detection[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). Piscataway, NJ, USA: IEEE, 2015: 25-32.
9	Song Hangke, Liu Zhi, Xie Yufeng, et al. RGBD Co-saliency Detection Via Bagging-based Clustering[J]. IEEE Signal Processing Letters, 2016, 23(12): 1722- 1726.
10	Cong Runmin, Lei Jianjun, Zhang Changqing, et al. Saliency Detection for Stereoscopic Images Based on Depth Confidence Analysis and Multiple Cues Fusion[J]. IEEE Signal Processing Letters, 2016, 23(6): 819-823.
11	Zhu Chunbiao, Li Ge, Wang Wenmin, et al. An Innovative Salient Object Detection Using Center-dark Channel Prior[C]//2017 IEEE International Conference on Computer Vision Workshops (ICCVW). Piscataway, NJ, USA: IEEE, 2017: 1509-1515.
12	Qu Liangqiong, He Shengfeng, Zhang Jiawei, et al. RGBD Salient Object Detection via Deep Fusion[J]. IEEE Transactions on Image Processing, 2017, 26(5): 2274-2285.
13	Chen Hao, Li Youfu. Progressively Complementarity- aware Fusion Network for RGB-D Salient Object Detection[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2018: 3051-3060.
14	Liu Zhengyi, Shi Song, Duan Quntao, et al. Salient Object Detection for RGB-D Image by Single Stream Recurrent Convolution Neural Network[J]. Neurocomputing, 2019, 363: 46-57.
15	Zhao Jiaxing, Cao Yang, Fan Dengping, et al. Contrast Prior and Fluid Pyramid Integration for RGBD Salient Object Detection[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, NJ, USA: IEEE, 2019: 3922-3931.
16	Fu Keren, Fan Dengping, Ji Gepeng, et al. JL-DCF: Joint Learning and Densely-cooperative Fusion Framework for RGB-D Salient Object Detection[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, NJ, USA: IEEE, 2020: 3049-3059.
17	Fan Dengping, Lin Zheng, Zhang Zhao, et al. Rethinking RGB-D Salient Object Detection: Models, Data Sets, and Large-scale Benchmarks[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(5): 2075-2089.
18	Feng Guang, Meng Jinyu, Zhang Lihe, et al. Encoder Deep Interleaved Network with Multi-scale Aggregation for RGB-D Salient Object Detection[J]. Pattern Recognition, 2022, 128: 108666.
19	Zhang Miao, Ren Weisong, Yongri Piao, et al. Select, Supplement and Focus for RGB-D Saliency Detection[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, NJ, USA: IEEE, 2020: 3469-3478.
20	Liu Chang, Yang Gang, Wang Shuo, et al. TANet: Transformer-based Asymmetric Network for RGB-D Salient Object Detection[J]. IET Computer Vision, 2023, 17(4): 415-430.
21	Pang Youwei, Zhang Lihe, Zhao Xiaoqi, et al. Hierarchical Dynamic Filtering Network for RGB-D Salient Object Detection[C]//Computer Vision-ECCV 2020. Cham: Springer International Publishing, 2020: 235-252.
22	Gao Shanghua, Cheng Mingming, Zhao Kai, et al. Res2Net: A New Multi-scale Backbone Architecture[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021, 43(2): 652-662.
23	Zhou Tao, Fu Huazhu, Chen Geng, et al. Specificity-preserving RGB-D Saliency Detection[C]//2021 IEEE/CVF International Conference on Computer Vision (ICCV). Piscataway, NJ, USA: IEEE, 2021: 4661-4671.
24	Ronneberger O, Fischer P, Brox T. U-net: Convolutional Networks for Biomedical Image Segmentation[C]// Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015. Cham: Springer International Publishing, 2015: 234-241.
25	Woo S, Park J, Lee J Y, et al. CBAM: Convolutional Block Attention Module[C]//Computer Vision-ECCV 2018. Cham: Springer International Publishing, 2018: 3-19.
26	Shi Xingjian, Chen Zhourong, Wang Hao, et al. Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting[C]//Proceedings of the 28th International Conference on Neural Information Processing Systems. Cambridge, MA, USA: MIT Press, 2015: 802-810.
27	Wei Jun, Wang Shuhui, Huang Qingming. F³Net: Fusion, Feedback and Focus for Salient Object Detection[C]//Proceedings of the AAAI Conference on Artificial Intelligence. Palo Alto, CA, USA: AAAI Press, 2020: 12321-12328.
28	Li Nianyi, Ye Jinwei, Ji Yu, et al. Saliency Detection on Light Field[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(8): 1605-1616.
29	Ju Ran, Liu Yang, Ren Tongwei, et al. Depth-aware Salient Object Detection Using Anisotropic Center-surround Difference[J]. Signal Processing: Image Communication, 2015, 38: 115-126.
30	Peng Houwen, Li Bing, Xiong Weihua, et al. RGBD Salient Object Detection: A Benchmark and Algorithms[C]//Computer Vision-ECCV 2014. Cham: Springer International Publishing, 2014: 92-109.
31	Niu Yuzhen, Geng Yujie, Li Xueqing, et al. Leveraging Stereopsis for Saliency Analysis[C]//2012 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2012: 454-461.
32	Margolin R, Zelnik-Manor L, Tal A. How to Evaluate Foreground Maps[C]//2014 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2014: 248-255.
33	Fan Dengping, Gong Cheng, Cao Yang, et al. Enhanced-alignment Measure for Binary Foreground Map Evaluation[C]///Proceedings of the 27th International Joint Conference on Artificial Intelligence. Palo Alto, CA, USA: AAAI Press, 2018: 698-704.
34	Achanta R, Sheila Hemami, Estrada F, et al. Frequency-tuned Salient Region Detection[C]//2009 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2009: 1597-1604.
35	Russakovsky O, Deng Jia, Su Hao, et al. ImageNet Large Scale Visual Recognition Challenge[J]. International Journal of Computer Vision, 2015, 115(3): 211-252.
36	Fu Keren, Fan Dengping, Ji Gepeng, et al. Siamese Network for RGB-D Salient Object Detection and Beyond[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(9): 5541-5559.
37	Li Gongyang, Liu Zhi, Chen Minyu, et al. Hierarchical Alternate Interaction Network for RGB-D Salient Object Detection[J]. IEEE Transactions on Image Processing, 2021, 30: 3528-3542.
38	Jin Wenda, Xu Jun, Han Qi, et al. CDNet: Complementary Depth Network for RGB-D Salient Object Detection[J]. IEEE Transactions on Image Processing, 2021, 30: 3376-3390.
39	Sun Peng, Zhang Wenhu, Wang Huanyu, et al. Deep RGB-D Saliency Detection with Depth-sensitive Attention and Automatic Multi-modal Fusion[C]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, NJ, USA: IEEE, 2021: 1407-1417.
40	Li Jingjing, Ji Wei, Bi Qi, et al. Joint Semantic Mining for Weakly Supervised RGB-D Salient Object Detection[C]//Advances in Neural Information Processing Systems. San Francisco, CA, USA: Curran Associates, Inc., 2021: 11945-11959.
41	Wu Yuhuan, Liu Yun, Xu Jun, et al. MobileSal: Extremely Efficient RGB-D Salient Object Detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(12): 10261-10269.
42	Ji Wei, Li Jingjing, Bi Qi, et al. Promoting Saliency from Depth: Deep Unsupervised RGB-D Saliency Detection[EB/OL]. (2022-05-15) [2022-08-31]. .
43	Chen Tianyou, Hu Xiaoguang, Xiao Jin, et al. CFIDNet: Cascaded Feature Interaction Decoder for RGB-D Salient Object Detection[J]. Neural Computing and Applications, 2022, 34(10): 7547-7563.
44	Wang Fengyun, Pan Jinshan, Xu Shoukun, et al. Learning Discriminative Cross-modality Features for RGB-D Saliency Detection[J]. IEEE Transactions on Image Processing, 2022, 31: 1285-1297.
45	Zhao Xiaoqi, Pang Youwei, Zhang Lihe, et al. Self-Supervised Pretraining for RGB-D Salient Object Detection[C]//Proceedings of the AAAI Conference on Artificial Intelligence. Palo Alto, CA, USA: AAAI Press, 2022: 3463-3471.
46	Zhang Miao, Yao Shunyu, Hu Beiqi, et al. C² DFNet: Criss-cross Dynamic Filter Network for RGB-D Salient Object Detection[J/OL]. IEEE Transactions on Multimedia, 2022: 1-13. (2022-07-01) [2022-09-05]. .
47	Simonyan K, Zisserman A. Very Deep Convolutional Networks for Large-scale Image Recognition[EB/OL]. (2015-04-10) [2022-09-11]. .

[1]	于豪, 蒋锦霞, 赖晓翰, 梅峰, 王庆. 基于自适应感受野的电力设备表面缺陷检测方法[J]. 系统仿真学报, 2023, 35(7): 1572-1580.
[2]	魏赟, 罗琦, 赵迎志. 基于自适应融合和注意力细化的语义分割模型[J]. 系统仿真学报, 2023, 35(6): 1226-1234.
[3]	石鼎, 燕雪峰, 宫丽娜, 张静宣, 关东海, 魏明强. 强化学习驱动的海战场多智能体协同作战仿真算法[J]. 系统仿真学报, 2023, 35(4): 786-796.
[4]	向南, 王璐, 贾崇柳, 蹇越谋, 马小霞. 改进YOLO的遮挡行人检测仿真[J]. 系统仿真学报, 2023, 35(2): 286-299.
[5]	孙红, 张玉香, 凌岳览. 基于损失提取反馈注意网络的图像超分辨率重建研究[J]. 系统仿真学报, 2023, 35(2): 308-317.
[6]	金炜东, 张述礼, 唐鹏, 张曼. 基于稠密残差块与通道像素注意力的图像去雾网络[J]. 系统仿真学报, 2022, 34(8): 1663-1673.
[7]	邱俊杰, 郑红, 程云辉. 基于多尺度LSTM预测模型研究[J]. 系统仿真学报, 2022, 34(7): 1593-1604.
[8]	王银, 王飞翔, 孙前来. 多尺度特征融合车辆检测方法[J]. 系统仿真学报, 2022, 34(6): 1219-1229.
[9]	尹诗, 侯国莲, 迟岩, 弓林娟, 胡晓东. 风电机组发电机前轴承健康度预测方法及实现[J]. 系统仿真学报, 2021, 33(6): 1323-1333.
[10]	徐胜, 冯文宇, 刘志诚, 涂鑫涛, 费敏锐, 张堃. 基于机器视觉的复杂环境下精确手势识别算法研究[J]. 系统仿真学报, 2021, 33(10): 2460-2469.
[11]	姜明新, 潘志庚, 王兰芳, 胡铸鑫. 基于深度去噪自编码器的RGB-D视频目标跟踪[J]. 系统仿真学报, 2018, 30(11): 4276-4283.
[12]	孔亦思, 胡晓峰, 朱丰, 陶九阳. 战场态势感知中的注意力机制探析[J]. 系统仿真学报, 2017, 29(10): 2233-2241.
[13]	蔡强, 魏立伟, 李海生, 曹健. 基于ANNet网络的RGB-D图像的目标检测[J]. 系统仿真学报, 2016, 28(9): 2260-2266.