Deep learning and optimization algorithm for high efficient searching and detection of aircraft targets in remote sensing images
-
摘要:
为了实现大幅面遥感图像中飞机目标的高效检测与准确定位，通过深度神经网络（DNN）的级联组合，提出了一种新颖的搜寻与检测相集成的飞机目标高效检测算法。首先，运用高性能的端到端DNN网络，进行停机坪与跑道区域的像素级高效精准分割，从而大幅度缩小飞机目标的搜索范围，以降低虚警发生概率，完成飞机目标候选检测区域的快速搜寻。然后，针对分割所得停机坪与跑道区域，借助手工数据集对YOLO网络模型进行迁移式强化训练，一方面可以弥补训练集在样本类型与数据规模上的不足，另一方面借助YOLO网络的强时效性优势对飞机目标的位置进行回归求解，可以显著提高飞机目标的检测效率。停机坪与跑道区域分割DNN网络在分割精度与时效性上具有显著优势，而迁移式强化训练YOLO网络不仅具有很高的检测效率，在检测精度上也能保持良好的性能。通过一系列综合实验与对比分析，验证了提出的搜寻与检测相集成的DNN级联组合式飞机目标高效检测算法的性能优势。
Abstract:In order to achieve high-performance detection and accurate positioning of aircraft targets in large-scale remote sensing images, in this paper, a kind of efficient aircraft target detection algorithm based on synthetic integration of searching and detection is presented. First, through the end-to-end deep neural networks (DNN), the efficient and accurate pixel-level segmentation of apron and runway area is achieved so that the searching range of aircraft targets is greatly narrowed and the probability of false alarm is also reduced effectively and the goal of high speed searching of aircraft targets candidate detection areas is achieved accordingly. In view of the segmented areas of apron and runway, the strategy of transfer reinforcement learning is employed to pre-trained YOLO networks with supervised signals of positive datasets by manual labelling. In this way, pre-trained networks can make up the deficiency of capacity of manual data sets, and the advantage of real-time property of YOLO networks can also be utilized to deal with the classification and locations of aircraft targets so as to achieve a satisfied solution of regression problems and promote the efficiency of detection significantly. It is obvious that the apron and runway segmentation with DNN networks can play important role in getting performance superiority of high precision and efficiency. Meanwhile, YOLO networks based on transfer reinforcement learning not only possess the characteristics of high efficiency, but also maintain the precision of detection at a high level. A series of comprehensive experiments and comparative analyses verify the effectiveness and good performance of the proposed searching and detection algorithm of aircraft targets with DNN cascade combination and synthetic integration.
-
图 1 不同区域停靠的飞机目标样本示意图
Figure 1. Schematic of aircraft target sample docked in various areas
图 2 停机坪与跑道区域分割DNN网络模型的离线监督训练
Figure 2. Off-line supervised training of DNN model for apron and runway area segmentation
图 4 停机坪与跑道区域分割DNN网络模型示意图
Figure 4. Schematic of DNN models for apron and runway area segmentation
图 6 停机坪与跑道区域分割DNN网络模型性能进化曲线
Figure 6. Performance evolution curves of DNN models for apron and runway area segmentation
图 7 不同DNN网络模型停机坪与跑道区域分割结果对比
Figure 7. Comparison of apron and runway area segmentation results among various DNN models
图 8 不同DNN网络模型停机坪与跑道区域分割IoU对比
Figure 8. Comparison of IoU for apron and runway area segmentation with various DNN models
图 9 停机坪与跑道区域分割时间开销对比
Figure 9. Comparison of time cost for apron and runway segmentation
表 1 不同区域停靠的飞机目标样本数量统计
Table 1. Quantity statistics of aircraft target sample docked in various areas
区域 飞机目标样本数量/架 比例/% 停机坪 1035 96.82 跑道 11 1.03 其他区域 23 2.15 
下载: 导出CSV
表 2 不同DNN网络模型停机坪与跑道区域分割IoU得分
Table 2. IoU for apron and runway area segmentation with various DNN models
DNN网络 FCN-16 FCN-8 SegNet U-Net LS-RSIASNN IoU得分 0.1730 0.3840 0.6495 0.7243 0.7454
下载: 导出CSV
表 3 飞机目标平均检测时间开销对比
Table 3. Comparison of average time cost for aircraft target detection
s 算法 平均检测时间开销 R-CNN 2.0257 Faster R-CNN 0.0946 YOLO 0.0277
下载: 导出CSV
表 4 飞机目标检测性能对比
Table 4. Performance comparison of aircraft target detection
算法 图像类型 精确率/% 漏检率/% 时间开销/s R-CNN 原始遥感图像 56.25 18.72 385.51 停机坪与跑道区域分割图像 78.32 15.35 87.49 Faster R-CNN 原始遥感图像 68.87 6.31 039.46 停机坪与跑道区域分割图像 94.69 5.19 28.29 YOLO 原始遥感图像 62.56 10.83 006.65 停机坪与跑道区域分割图像 90.28 9.26 4.76
下载: 导出CSV
-
[1] HAN Z X, ZHANG H, ZHANG J F, et al.Fast aircraft detection based on region locating network in large-scale remote sensing images[C]//24th IEEE International Conference on Image Processing.Piscataway, NJ: IEEE Press, 2017: 2294-2298. [2] GIRSHICK R, DONAHUE J, DARRELL T, et al.Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Piscataway, NJ: IEEE Press, 2014: 580-587. [3] REN S, HE K, GIRSHICK R, et al.Faster R-CNN:Towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6):1137-1149. doi: 10.1109/TPAMI.2016.2577031 [4] 赵雪梅, 李玉, 赵泉华.基于隐马尔可夫高斯随机场模型的模糊聚类高分辨率遥感影像分割算法[J].电子学报, 2016, 44(3):679-686. doi: 10.3969/j.issn.0372-2112.2016.03.028ZHAO X M, LI Y, ZHAO Q H.Hidden Markov Gaussian random field based fuzzy clustering algorithm for high-resolution remote sensing image segmentation[J]. Acta Electronica Sinica, 2016, 44(3):679-686(in Chinese). doi: 10.3969/j.issn.0372-2112.2016.03.028 [5] 陈荣元, 徐雪松, 李广琼, 等.自适应特征加权的Gibbs随机场影像分割方法[J].电子学报, 2016, 44(10):2351-2356. doi: 10.3969/j.issn.0372-2112.2016.10.010CHEN R Y, XU X S, LI G Q, et al.Image segmentation by combining adaptively weighted features with Gibbs random field[J]. Acta Electronica Sinica, 2016, 44(10):2351-2356(in Chinese). doi: 10.3969/j.issn.0372-2112.2016.10.010 [6] BUDAK U, HALICI U, SENGUR A, et al.Efficient airport detection using line segment detector and Fisher vector representation[J]. IEEE Geoscience & Remote Sensing Letters, 2016, 13(8):1079-1083. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=8b90b1ce53fa228655d3e4964573adf7 [7] WANG Y, PAN L.Automatic airport recognition based on saliency detection and semantic information[J]. ISPRS International Journal of Geo-Information, 2016, 5(7):115-118. doi: 10.3390/ijgi5070115 [8] KRIZHEVSKY A, SUTSKEVER I, HINTON G E.ImageNet classification with deep convolutional neural networks[C]//International Conference on Neural Information Processing Systems, 2012: 1097-1105. [9] LONG J, SHELHAMER E, DARRELL T.Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4):640-651. doi: 10.1109/TPAMI.2016.2572683 [10] BADRINARAYANAN V, KENDALL A, CIPOLLA R.SegNet:A deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(12):2481-2495. doi: 10.1109/TPAMI.2016.2644615 [11] RONNEBERGER O, FISCHER P, BROX T.U-Net: Convolutional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Computer-assisted Intervention.Berlin: Springer, 2015: 234-241. [12] DAN C C, GIUSTI A, GAMBARDELLA L M, et al.Deep neural networks segment neuronal membranes in electron microscopy images[J]. Advances in Neural Information Processing Systems, 2012, 25:2852-2860. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=CC0213884171 [13] CHEN L, PAPANDREOU G, KOKKINOS I, et al.Semantic image segmentation with deep convolutional nets and fully connected CRFs[J]. Computer Science, 2014(4):357-361. http://d.old.wanfangdata.com.cn/Periodical/jsjyy201806005 [14] ZHAO H, SHI J, QI X, et al.Pyramid scene parsing network[C]//IEEE Conference on Computer Vision and Pattern Recognition.Piscataway, NJ: IEEE Press, 2017: 6230-6239. [15] CHEN L C, PAPANDREOU G, KOKKINOS I, et al.DeepLab:Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFS[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(4):834-848. doi: 10.1109/TPAMI.2017.2699184 [16] HE K, GKIOXARI G, DOLLAR P, et al.Mask R-CNN[C]//IEEE International Conference on Computer Vision.Piscataway, NJ: IEEE Press, 2017: 2980-2988. [17] GIRSHICK R.Fast R-CNN[C]//IEEE International Conference on Computer Vision.Piscataway, NJ: IEEE Press, 2015: 1440-1448. [18] HE K, ZHANG X, REN S, et al.Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(9):1904-1916. doi: 10.1109/TPAMI.2015.2389824 [19] REDMON J, DIVVALA S, GIRSHICK R, et al.You only look once: Unified, real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Piscataway, NJ: IEEE Press, 2016: 779-788. [20] TORRALBA A, RUSSELL B C, YUEN J.LabelMe:Online image annotation and applications[J]. Proceedings of the IEEE, 2010, 98(8):1467-1484. doi: 10.1109/JPROC.2010.2050290 [21] HE K, ZHANG X, REN S, et al.Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Piscataway, NJ: IEEE Press, 2016: 770-778. [22] HUANG G, LIU Z, WEINBERGER K Q, et al.Densely connected convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Piscataway, NJ: IEEE Press, 2017: 3-8. [23] SU W, WANG Z.Widening residual skipped network for semantic segmentation[J]. IET Image Processing, 2017, 11(10):880-887. doi: 10.1049/iet-ipr.2017.0070 [24] KINGMA D P, BA J.Adam: A method for stochastic optimization[EB/OL]. (2014-12-22).https://arxiv.org/abs/1412.6980. -
下载:
点击查看大图
计量
- 文章访问数: 645
- HTML全文浏览量: 116
- PDF下载量: 528
- 被引次数: 0
