[1]唐桂荣,许 健,钱苏平,等.无人机河湖巡检中RCNN识别算法应用研究[J].江苏水利,2021,(01):34-38.
　TANG Guirong,XU Jian,QIAN Suping,et al.Research on application of RCNN identification algorithm in UAVS river and lake patrol inspection[J].JIANGSU WATER RESOURCES,2021,(01):34-38.
点击复制

无人机河湖巡检中RCNN识别算法应用研究()
分享到：

《江苏水利》[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2021年01期
页码:
34-38
栏目:
水利工程建设
出版日期:
2021-02-03

文章信息/Info

Title:
Research on application of RCNN identification algorithm in UAVS river and lake patrol inspection
文章编号:
1007-7839(2021)01-0034-05
作者:
唐桂荣1;  许 健2;  钱苏平1;  付 港3;  钱晓军3
1.泰州市水利局, 江苏 泰州 225300; 2.泰州姜堰区水利局, 江苏 泰州 225500; 3.南京师范大学计算机科学与技术学院, 江苏 南京 210023
Author(s):
TANG Guirong1;  XU Jian2;  QIAN Suping1;  FU Gang3;  QIAN Xiaojun3
1.Taizhou Water Resources Bureau, Taizhou 225300, China; 2.Taizhou Jiangyan Water Resources Bureau, Taizhou 225500, China; 3.School of Computer and Electronic Information, Nanjing Normal University,Nanjing 210023, China
关键词:
河湖治理;  RCNN;  数字图像处理
Keywords:
river and lake management;  RCNN;  digital image processing
分类号:
TV85
文献标志码:
B
摘要:
随着经济发展与人口增长,带来了河湖治理难度增加、水体污染频发等问题,针对这些情况设计出一种基于区域卷积神经网络算法(RCNN)的无人机河湖巡检图像智能处理平台。通过聚类和构建卷积神经网络提取可能含有河湖污染的区域,利用RCNN对目标区域进行分类和识别,以此作为问题源构建河湖管理的闭环,较好的完成巡检目标。
Abstract:
With economic development and population growth, problems such as increasing difficulty in river and lake control and frequent water pollution have been brought about. In response to these situations, an intelligent processing platform for UAV river and lake inspection images based on regional convolutional neural network algorithm(RCNN)was designed. Through clustering and constructing a convolutional neural network, areas that might contain river and lake pollution were extracted, and RCNN was used to classify and identify the target area, which was used as the source of the problem to construct a closed loop of river and lake management, and to better complete the inspection target.

参考文献/References:

[1] 刘琼, 李宗贤, 孙富春, 等. 基于深度信念卷积神经网络的图像识别与分类[J]. 清华大学学报(自然科学版), 2018, 58(9):781-787.
[2] 徐露露. 基于深度卷积特征的迁移学习在图像识别上的应用研究[D]. 广州:华南理工大学, 2018.
[3] MAHMOOD Z, SAFDER I, NAWAB RMA, et al. Deep sentiments in Roman Urdu text using recurrent convolutional neural network model[J]. Information Processing and Management, 2020, 57(4):25-27.
[4] PIERDICCA R, PAOLANTI M, NASPETTI S, et al. User-centered predictive model for improving cultural heritage augmented reality applications: an HMM-based approach for eye-tracking data[J]. Journal of Imaging, 2018, 4(8):101.
[5] 周世兵, 徐振源, 唐旭清. K-means算法最佳聚类数确定方法[J]. 计算机应用, 2010, 30(8):1995-1998.
[6] 温尧乐, 李林燕, 尚欣茹, 等. 一种改进的Mask RCNN特征融合实例分割方法[J]. 计算机应用与软件,2019, 36(10):130-133.
[7] 赵文清, 程幸福, 赵振兵, 等. 注意力机制和Faster RCNN相结合的绝缘子识别[J/OL]. 智能系统学报:1-7[2020-04-01].
[8] BA ROM KANG, HYUNKU LEE, KEUNJU PRK, et al. BshapeNet:Object detection and instance segmentation with bounding shape masks[J]. Pattern Recognition Letters, 2020(131):1.

相似文献/References:

[1]蒋志昊,梁文广,祁仰旭,等.江苏省省级河长制河湖“三乱”治理流程初探[J].江苏水利,2020,(06):58.
　JIANG Zhihao,LIANG Wenguang,QI Yangxu,et al.Preliminary discussion on the governance process of "three chaos" in river and lake of provincial system of river leader in Jiangsu[J].JIANGSU WATER RESOURCES,2020,(01):58.
[2]韩全林.赓续使命 坚毅笃行 不断开创河湖管理保护新局面[J].江苏水利,2022,(增刊1):28.
　HAN Quanlin.Continue to carry out our mission with perseverance, and continue to break new ground in rivers and lakes management and protection[J].JIANGSU WATER RESOURCES,2022,(01):28.
[3]郑露,刘鹏,唐仁,等.洮滆地区河湖生态治理路径探索[J].江苏水利,2024,(03):17.
　ZHENG Lu,LIU Peng,TANG Ren,et al.Exploration on ecological governance path for rivers and lakes in the Taoge Region[J].JIANGSU WATER RESOURCES,2024,(01):17.

备注/Memo

备注/Memo:
收稿日期:2020-01-06
基金项目:江苏省水利科技项目(2019052)
作者简介:唐桂荣(1978—),工程师,主要从事河湖治理、河长制管理工作。E-mail:874783521@qq.com
更新日期/Last Update: 2021-01-20