一种基于可见光巡检图像的杂草智能识别方法

doi:10.11930/j.issn.1004-9649.201902152

中国电力 ›› 2019, Vol. 52 ›› Issue (11): 138-144,174.DOI: 10.11930/j.issn.1004-9649.201902152

一种基于可见光巡检图像的杂草智能识别方法

岳国良¹, 路艳巧², 常浩³, 孙翠英²

1. 国网河北省电力有限公司, 河北石家庄 050041;
2. 国网河北省电力有限公司电力科学研究院, 河北石家庄 050021;
3. 国网河北省电力有限公司检修分公司, 河北石家庄 050070

收稿日期:2019-02-28 修回日期:2019-05-07 出版日期:2019-11-05 发布日期:2019-11-05
通讯作者: 路艳巧(1986-),女,通信作者,高级工程师,从事电网设备带电检测和在线监测新技术研究,E-mail:2252105702@qq.com
作者简介:岳国良(1971-),男,博士,高级工程师(教授级),从事状态检修、智能运维检测领域的研究,E-mail:18003218219@189.cn
基金资助:
国网河北省电力有限公司电力科学研究院项目（基于深度学习技术的电网设备图像识别与故障检测技术研究，KJKF-20）

An Intelligent Weed Recognition Method Based on Optical Patrol Image

YUE Guoliang¹, LU Yanqiao², CHANG Hao³, SUN Cuiying²

1. State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050041, China;
2. State Grid Hebei Electric Power Research Institute, Shijiazhuang 050021, China;
3. Maintenance Branch of State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050070, China

Received:2019-02-28 Revised:2019-05-07 Online:2019-11-05 Published:2019-11-05
Supported by:
This work is supported by Program of State Grid Hebei Electric Power Research Institute (Research of Image Recognition and Fault Detection Technique for Grid Equipment Based on Deep Learning Techniques, No.KJKF-20)

摘要/Abstract

摘要： 目前电力巡检主要是采用无人机巡检的方式，针对无人机巡检获取的图像识别过程中，电力设备旁的杂草可能会造成安全隐患，需要对图像中的杂草进行识别。针对电力巡检的场景，提出了一种基于可见光巡检图像的杂草智能识别方法，以可见光巡检图像中杂草的特征为基础，结合卷积神经网络方法，解决可见光巡检图像中电力设备附近的杂草识别问题。通过对图像进行样本数据增广和预处理，接着引入区域生成网络，再对图像提取固定个数候选框的图像特征，和改进的图像分类网络连接在一起，得到最终的卷积神经网络模型。实验表明其准确率可以达到97.98%，检测一幅600×600大小图像需要花费的平均时间约为0.256 s，在保证了准确率的同时达到了高效识别的要求。

关键词: 巡检图像, 卷积神经网络, 区域生成, 图像分类, 杂草识别, 人工智能与大数据应用

Abstract: Power inspection is currently carried out mainly by drones. The weeds around the power equipment may cause potential safety hazards when the patrol images acquired by drones are used for patrol inspection, it is therefore necessary to recognize the weeds in the image. In this paper, a method for intelligent recognition of weeds is proposed for power patrol inspection based on optical patrol images. Based on the feature of weeds in the optical images, and combined with the convolutional neural network method, the problem of weed recognition near the power equipment in optical patrol images is solved. By amplifying and preprocessing the sample data of the optical patrol images, and introducing the region proposal network, the image features of the fixed number of candidate frames are extracted from the images. Then the network is connected to the improved image classification network to obtain a final convolutional neural network model. The experiments show that the accuracy rate can reach 97.98%, and the average time taken for detecting a 600×600 image is around 0.256 seconds, which meets the requirements of efficient recognition while ensuring accuracy.

Key words: patrol image, convolutional neural network, region proposal, image classification, weed recognition, artificial intelligence and big data application

中图分类号:

TM762
TP391

岳国良, 路艳巧, 常浩, 孙翠英. 一种基于可见光巡检图像的杂草智能识别方法[J]. 中国电力, 2019, 52(11): 138-144,174.

YUE Guoliang, LU Yanqiao, CHANG Hao, SUN Cuiying. An Intelligent Weed Recognition Method Based on Optical Patrol Image[J]. Electric Power, 2019, 52(11): 138-144,174.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.201902152

https://www.electricpower.com.cn/CN/Y2019/V52/I11/138

参考文献

[1] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]//2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Washington D.C.:IEEE Computer Society, 2005:886-893.
[2] FREUND Y, SCHAPIRE R E. A decision-theoretic generalization of on-line learning and an application to boosting[J]. Journal of Computer and System Sciences, 1997, 55(1):119-139.
[3] VIOLA P, JONES M. Rapid object detection using a boosted cascade of simple features[C]//Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Washington D.C.:IEEE Computer Society, 2001.
[4] FELZENSZWALB P F, GIRSHICK R B, MCALLESTER D. Cascade object detection with deformable part models[C]//2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Washington D.C.:IEEE Computer Society, 2010:2241-2248.
[5] EVERINGHAM M, VAN GOOL L, WILLIAMS C K I, et al. The pascal visual object classes (VOC) challenge[J]. International Journal of Computer Vision, 2010, 88(2):303-338.
[6] ZITNICK C L, DOLLÁR P. Edge boxes:locating object proposals from edges[C]//European Conference on Computer Vision-ECCV 2014, Cham:Springer, 2014:391-405.
[7] UIJLINGS J R R, VAN DE SANDE K E A, GEVERS T, et al. Selective search for object recognition[J]. International Journal of Computer Vision, 2013, 104(2):154-171.
[8] 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.
[9] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once:unified, real-time object detection[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition, Washington D.C.:IEEE Computer Society, 2016:779-788.
[10] LIU W, ANGUELOV D, ERHAN D, et al. SSD:single shot multibox detector[C]//European Conference on Computer Vision-ECCV 2016, Cham:Springer, 2016:21-37.
[11] ZHANG X, ZHOU X, LIN M, et al. Shufflenet:an extremely efficient convolutional neural network for mobile devices[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Washington D.C.:IEEE Computer Society, 2018:6848-6856.
[12] CHOLLET F. Xception:deep learning with depthwise separable convolutions[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition, Washington D.C.:IEEE Computer Society, 2017:1800-1807.
[13] QI S, MA J, LIN J, et al. Unsupervised ship detection based on saliency and S-HOG descriptor from optical satellite images[J]. IEEE Geoscience and Remote Sensing Letters, 2015, 12(7):1451-1455.
[14] CORBANE C, NAJMAN L, PECOUL E, et al. A complete processing chain for ship detection using optical satellite imagery[J]. International Journal of Remote Sensing, 2010, 31(22):5837-5854.
[15] ZHU C, ZHOU H, WANG R, et al. A novel hierarchical method of ship detection from spaceborne optical image based on shape and texture features[J]. IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(9):3446-3456.
[16] BI F, ZHU B, GAO L, et al. A visual search inspired computational model for ship detection in optical satellite images[J]. IEEE Geoscience and Remote Sensing Letters, 2012, 9(4):749-753.
[17] 孙俊, 何小飞, 谭文军, 等. 空洞卷积结合全局池化的卷积神经网络识别作物幼苗与杂草[J]. 农业工程学报, 2018, 34(11):159-165 SUN Jun, HE Xiaofei, TAN Wenjun, et al. Recognition of crop seedling and weed recognition based on dilated convolution and global pooling in CNN[J]. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(11):159-165
[18] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Imagenet classification with deep convolutional neural networks:advances in neural information processing systems[C]//2012 TEEE Computer Society Conference on Computer Vision and Pattern Recognition, Washington D.C.:IEEE Computer Society, 2012:1097-1105.
[19] 王璨, 武新惠, 李志伟. 基于卷积神经网络提取多尺度分层特征识别玉米杂草[J]. 农业工程学报, 2018, 34(5):144-151 WANG Can, WU Xinhui, LI Zhiwei. Recognition of maize and weed based on multi-scale hierarchical features extracted by convolutional neural network[J]. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(5):144-151
[20] 姜红花, 王鹏飞, 张昭, 等. 基于卷积网络和哈希码的玉米田间杂草快速识别方法[J]. 农业机械学报, 2018, 49(11):30-38 JIANG Honghua, WANG Pengfei, ZHANG Zhao, et al. Fast Identification of field weeds based on deep convolutional network and binary hash code[J]. Transactions of the Chinese Society for Agricultural Machinery, 2018, 49(11):30-38
[21] 张旭, 王斌, 张旭. CNN在农田杂草图像识别的可行性探讨[J]. 电脑知识与技术, 2018, 14(22):187-189 ZHANG Xu, WANG Bin, ZHANG Xu. Discussion on feasibility of CNN with image recognition in farmland weeds[J]. Computer Knowledge and Technology, 2018, 14(22):187-189

一种基于可见光巡检图像的杂草智能识别方法

An Intelligent Weed Recognition Method Based on Optical Patrol Image

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	吴军英, 路欣, 刘宏, 张彬, 柴守亮, 刘蕴春, 王佳楠. 基于Spearman-GCN-GRU模型的超短期多区域电力负荷预测[J]. 中国电力, 2024, 57(6): 131-140.
[2]	娄奇鹤, 李荣盛, 谭捷, 袁铁江. 基于卷积神经网络的暂稳极限功率计算[J]. 中国电力, 2024, 57(4): 211-219.
[3]	周颖, 白雪峰, 王阳, 邱敏, 孙冲, 武亚杰, 李彬. 面向虚拟电厂运营的温度敏感负荷分析与演变趋势研判[J]. 中国电力, 2024, 57(1): 9-17.
[4]	王大兴, 宁妍, 汪敬培, 徐洋, 毕峻, 周铭标, 王鹏. 构建新型电力系统背景下的微电网鲁棒简化建模[J]. 中国电力, 2024, 57(1): 148-157.
[5]	皮志勇, 朱益, 廖玄, 李振兴, 方豪, 吴沛. 基于深度学习的智能变电站通信链路故障定位方法[J]. 中国电力, 2023, 56(7): 136-145.
[6]	陆友文, 崔昊, 陈佳宁, 彭祥佳, 冯双, 刘栋. 基于RA-CNN和同步相量的风电场次/超同步振荡参数智能辨识方法[J]. 中国电力, 2023, 56(4): 46-55,67.
[7]	武小琳, 栾凌, 潘连武, 李海龙. 基于LM-CNN的输变电工程造价自动计算模型[J]. 中国电力, 2023, 56(2): 157-163.
[8]	陈子含, 滕伟, 胥学峰, 丁显, 柳亦兵. 基于图卷积网络和风速差分拟合的中长期风功率预测[J]. 中国电力, 2023, 56(10): 96-105.
[9]	冯裕祺, 李辉, 李利娟, 周彦博, 谭貌, 彭寒梅. 基于CNN-GRU的光伏电站电压轨迹预测[J]. 中国电力, 2022, 55(7): 163-171.
[10]	沙骏, 徐雨森, 刘冲冲, 冯定东, 胥峥, 臧海祥. 基于变分模态分解和分位数卷积-循环神经网络的短期风功率预测[J]. 中国电力, 2022, 55(12): 61-68.
[11]	樊江川, 于昊正, 刘慧婷, 杨丽君, 安佳坤. 基于多分支门控残差卷积神经网络的短期电力负荷预测[J]. 中国电力, 2022, 55(11): 155-162,174.
[12]	杨胡萍, 余阳, 汪超, 李向军, 胡奕涛, 饶楚楚. 基于VMD-CNN-BIGRU的电力系统短期负荷预测[J]. 中国电力, 2022, 55(10): 71-76.
[13]	曾囿钧, 肖先勇, 徐方维, 郑林. 基于CNN-BiGRU-NN模型的短期负荷预测方法[J]. 中国电力, 2021, 54(9): 17-23.
[14]	徐浩, 刘利强, 吕超. 基于迁移学习的配电网内部过电压识别方法[J]. 中国电力, 2021, 54(8): 52-59.
[15]	白万荣, 张驯, 朱小琴, 刘吉祥, 程其玉, 赵琰, 邵洁. 基于E-FCNN的电力巡检图像增强[J]. 中国电力, 2021, 54(5): 179-185.

模态框（Modal）标题

一种基于可见光巡检图像的杂草智能识别方法

An Intelligent Weed Recognition Method Based on Optical Patrol Image

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics