REMOTE SENSING TECHNOLOGY BASED ON IMPROVED FASTER R-CNN ALGORITHM AND ITS APPLICATION IN GEOLOGICAL DISASTER MONITORING
-
摘要:
为了减少地质灾害带给人类的破坏, 对自然环境中的地质信息进行精确监测具有极为重要的意义. 在对地质灾害和遥感技术之间的关联性进行描述的基础上, 本研究引入Faster R-CNN算法, 进一步运用特征提取网络, 修改训练方法并对该算法进行优化, 使之应用于遥感技术以提升检测地质灾害的速度和精度, 最终通过PASCAL VOC数据集与COCO数据集对所提出方法模型进行实验评价及论证. 结果发现, GIOU值为0.6时, 检测精度达到最优, 且在算法对比中, 本研究所设计的Faster R-CNN改进算法的AP@0.5达到了59, 证明该算法兼顾了速度与精确性, 达到了预期的目标检测目的. 同时, 发现本次研究设计的Faster R-CNN改进目标检测算法能够有效应用于卫星遥感技术, 达到快速检测到自然环境中的地质信息, 并能够通过判断其属性变换而做出预警措施, 从而使得人类因地质变化引起的损失得以减少.
-
关键词:
- 卫星遥感 /
- 地质灾害 /
- 环境监测 /
- 目标检测 /
- Faster R-CNN算法
Abstract:To reduce the damage caused by geological disasters to human beings, accurate monitoring of geological information in natural environment is of great significance. Based on the correlation between geological hazards and remote sensing technology, the study introduces the faster regions with convolutional neural network (Faster R-CNN) algorithm, further utilizes feature extraction network, modifies the training method and optimizes the algorithm, so that it can be applied to remote sensing technology to improve the speed and accuracy of geological hazards detection. Finally, the PASCAL VOC and COCO datasets are used for the experimental evaluation and demonstration of the proposed method model. The results show that when the GIOU value is 0.6, the detection accuracy reaches the best. In the algorithm comparison, the AP@0.5 value of the improved Faster R-CNN algorithm designed in the study reaches 59, proving that the algorithm is satisfactory in both speed and accuracy, and achieves the expected goal of target detection. Moreover, the Faster R-CNN algorithm can be effectively applied to satellite remote sensing technology to quickly detect geological information in natural environment and make early warning measures by judging its attribute transformation, thus reducing losses caused by geological changes.
-
-
[1] Meng R H, Rice S G, Wang J, et al. A fusion steganographic algorithm based on faster R-CNN[J]. Computers, Materials & Continua, 2018, 55(1): 1-16.
[2] 葛大庆. 地质灾害早期识别与监测预警中的综合遥感应用[J]. 城市与减灾, 2018(6): 53-60. https://www.cnki.com.cn/Article/CJFDTOTAL-CSJZ201806011.htm
Ge D Q. Comprehensive application of remote sensing in early identification, monitoring and early warning in geological disasters[J]. City and Disaster Reduction, 2018(6): 53-60. https://www.cnki.com.cn/Article/CJFDTOTAL-CSJZ201806011.htm
[3] 毕瑞, 甘淑, 李绕波, 等. 面向东川复杂山地泥石流沟谷三维地形建模及特征分析的无人机遥感探测应用研究[J]. 中国地质灾害与防治学报, 2021, 32(3): 91-100. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGDH202103012.htm
Bi R, Gan S, Li R B, et al. Application research of unmanned aerial vehicle remote sensing detection for 3D terrain modelling and feature analysis of debris flow gullies in complex mountainous area of Dongchuan[J]. The Chinese Journal of Geological Hazard and Control, 2021, 32(3): 91-100. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGDH202103012.htm
[4] Lian X G, Li Z J, Yuan H Y, et al. Rapid identification of landslide, collapse and crack based on low-altitude remote sensing image of UAV[J]. Journal of Mountain Science, 2020, 17(12): 2915-2928. doi: 10.1007/s11629-020-6080-9
[5] Athanassas C D. Muography for geological hazard assessment in the South Aegean active volcanic arc (SAAVA)[J]. Mediterranean Geoscience Reviews, 2020, 2(2): 233-246. doi: 10.1007/s42990-020-00020-x
[6] Lai F, Shao Q F, Lin Y, et al. A method for the hazard assessment of regional geological disasters: A case study of the Panxi area, China[J]. Journal of Spatial Science, 2021, 66(1): 143-162. doi: 10.1080/14498596.2019.1606741
[7] Qi L, Li B Y, Chen L K, et al. Ship target detection algorithm based on improved faster R-CNN[J]. Electronics, 2019, 8(9): 959. doi: 10.3390/electronics8090959
[8] Gao M Y, Du Y J, Yang Y X, et al. Adaptive anchor box mechanism to improve the accuracy in the object detection system[J]. Multimedia Tools and Applications, 2019, 78(19): 27383-27402. doi: 10.1007/s11042-019-07858-w
[9] Liu P, Wei Y M, Wang Q J, et al. A research on landslides automatic extraction model based on the improved mask R-CNN[J]. ISPRS International Journal of Geo-Information, 2021, 10(3): 168. doi: 10.3390/ijgi10030168
[10] Vabalas A, Gowen E, Poliakoff E, et al. Machine learning algorithm validation with a limited sample size[J]. PLoS One, 2019, 14(11): e0224365. doi: 10.1371/journal.pone.0224365
[11] Krinitskiy M, Aleksandrova M, Verezemskaya P, et al. On the generalization ability of data-driven models in the problem of total cloud cover retrieval[J]. Remote Sensing, 2021, 13(2): 326. doi: 10.3390/rs13020326
[12] 许强, 董秀军, 李为乐. 基于天-空-地一体化的重大地质灾害隐患早期识别与监测预警[J]. 武汉大学学报·信息科学版, 2019, 44(7): 957-966. https://www.cnki.com.cn/Article/CJFDTOTAL-WHCH201907002.htm
Xu Q, Dong X J, Li W L. Integrated space-air-ground early detection, monitoring and warning system for potential catastrophic geohazards[J]. Geomatics and Information Science of Wuhan University, 2019, 44(7): 957-966. https://www.cnki.com.cn/Article/CJFDTOTAL-WHCH201907002.htm
[13] Siregar S P, Wanto A. Analysis of artificial neural network accuracy using backpropagation algorithm in predicting process (forecasting)[J]. IJISTECH (International Journal of Information System and Technology), 2017, 1(1): 34-42.
[14] Xu Q H, Yuan S F, Huang T X. Multi-dimensional uniform initialization Gaussian mixture model for spar crack quantification under uncertainty[J]. Sensors, 2021, 21(4): 1283.
[15] 张春晓, 鲍云飞, 马中祺, 等. 基于卷积神经网络的光学遥感目标检测研究进展[J]. 航天返回与遥感, 2020, 41(6): 45-55. https://www.cnki.com.cn/Article/CJFDTOTAL-HFYG202006006.htm
Zhang C X, Bao Y F, Ma Z Q, et al. Research progress on optical remote sensing object detection based on CNN[J]. Spacecraft Recovery & Remote Sensing, 2020, 41(6): 45-55. https://www.cnki.com.cn/Article/CJFDTOTAL-HFYG202006006.htm
[16] Tanatipuknon A, Aimmanee P, Watanabe Y, et al. Study on combining two faster R-CNN models for landslide detection with a classification decision tree to improve the detection performance[J]. Journal of Disaster Research, 2021, 16(4): 588-595.
[17] Mahmood T, Arsalan M, Owais M, et al. Artificial intelligence-based mitosis detection in breast cancer histopathology images using faster R-CNN and deep CNNs[J]. Journal of Clinical Medicine, 2020, 9(3): 749.
[18] Zhang N, Feng Y R, Lee E J. Activity object detection based on improved faster R-CNN[J]. Journal of Korea Multimedia Society, 2021, 24(3): 416-422.
[19] Singh S, Ahuja U, Kumar M, et al. Face mask detection using YOLOv3 and faster R-CNN models: COVID-19 environment[J]. Multimedia Tools and Applications, 2021, 80(13): 19753-19768.
[20] Chen K, Tao W B. Learning linear regression via single-convolutional layer for visual object tracking[J]. IEEE Transactions on Multimedia, 2019, 21(1): 86-97.
[21] 刘小波, 刘鹏, 蔡之华, 等. 基于深度学习的光学遥感图像目标检测研究进展[J]. 自动化学报, 2021, 47(9): 2078-2089. https://www.cnki.com.cn/Article/CJFDTOTAL-MOTO202109002.htm
Liu X B, Liu P, Cai Z H, et al. Research progress of optical remote sensing image object detection based on deep learning[J]. Acta Automatica Sinica, 2021, 47(9): 2078-2089. https://www.cnki.com.cn/Article/CJFDTOTAL-MOTO202109002.htm
[22] Zhao K, Wang Y N, Zhu Q, et al. Intelligent detection of parcels based on improved faster R-CNN[J]. Applied Sciences, 2022, 12(14): 7158.
[23] Martín-Doñas J M, Gomez A M, Gonzalez J A, et al. A deep learning loss function based on the perceptual evaluation of the speech quality[J]. IEEE Signal Processing Letters, 2018, 25(11): 1680-1684.
[24] 欧攀, 张正, 路奎, 等. 基于卷积神经网络的遥感图像目标检测[J]. 激光与光电子学进展, 2019, 56(5): 051002. https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ201905009.htm
Ou P, Zhang Z, Lu K, et al. Object detection of remote sensing images based on convolutional neural networks[J]. Laser & Optoelectronics Progress, 2019, 56(5): 051002. https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ201905009.htm
-
下载:
图(7)
计量
- 文章访问数: 619
- PDF下载数: 22
- 施引文献: 0