基于RCNN的無人機巡檢圖像電力小部件識別研究
發(fā)布時間：2018-11-26 12:46
【摘要】：隨著無人機(UAV)在電力巡線作業(yè)中的應用推廣,對無人機巡檢圖像的信息挖掘或目標識別需求也越來越強烈。傳統(tǒng)的電力部件識別流程常使用經(jīng)典的機器學習算法,如支持向量機(SVM)、隨機森林或adaboost,結合梯度、顏色或紋理等淺層特征來對電力部件進行識別,難以充分利用無人機巡檢圖像的信息,并且難以達到較高的準確率。卷積神經(jīng)網(wǎng)絡(CNN)在目標識別中表現(xiàn)優(yōu)異,在很多目標識別場景之中成為首選算法。基于區(qū)域的卷積神經(jīng)網(wǎng)絡(RCNN)通過使用CNN從圖像中提取可能含有目標的區(qū)域來檢測并識別目標,但是計算復雜,難以滿足識別海量電力巡檢圖片的需求。Fast R-CNN和Faster RCNN利用CNN網(wǎng)絡提取圖像特征,后接一個區(qū)域提議層,優(yōu)化了提取可能含有目標區(qū)域的方式并改進識別目標的分類器,使目標的檢測和識別幾乎實時。本文詳細描述了Faster R-CNN算法流程,并在無人機電力線巡檢圖像部件檢測中使用,然后分別對DPM、SPPnet和Faster R-CNN識別方法進行了對比分析,利用實際采集的電力小部件巡檢數(shù)據(jù)構建的數(shù)據(jù)集對3種方法進行測試驗證,并討論了不同參數(shù)對識別結果的影響。實驗結果表明,基于深度學習的識別方法實現(xiàn)電力小部件的識別是可行的,而且利用Faster R-CNN進行多種類別的電力小部件識別定位可以達到每張近80 ms的識別速度和92.7%的準確率。
[Abstract]:With the application of UAV (UAV) in power line inspection, the demand of UAV patrol image information mining or target recognition is becoming more and more intense. Traditional power component recognition processes often use classical machine learning algorithms such as support vector machine (SVM) (SVM), random forest or adaboost, combined with gradient color or texture to identify power components. It is difficult to make full use of the image information of UAV patrol, and it is difficult to achieve high accuracy. Convolutional neural network (CNN) is the best algorithm for target recognition because of its excellent performance in target recognition. The region based convolution neural network (RCNN) detects and recognizes the target by using CNN to extract the region that may contain the target from the image, but the computation is complicated. Fast R-CNN and Faster RCNN use CNN network to extract image features, followed by a regional proposal layer, which optimizes the way of extracting possible target areas and improves the classifier for target recognition. The detection and recognition of target is almost in real time. This paper describes the flow of Faster R-CNN algorithm in detail, and uses it in the detection of UAV power line inspection image components, and then compares and analyzes the DPM,SPPnet and Faster R-CNN recognition methods, respectively. The three methods are tested and verified by the data set constructed from the actual data collected from the patrol inspection of power widget, and the influence of different parameters on the identification results is discussed. The experimental results show that the recognition method based on depth learning is feasible. Moreover, the recognition speed of 80 ms and the accuracy of 92.7% can be achieved by using Faster R-CNN to identify and locate various kinds of power components.
【作者單位】： 國網(wǎng)山東省電力公司電力科學研究院國家電網(wǎng)公司電力機器人技術實驗室;山東魯能智能技術有限公司;國網(wǎng)山東省電力公司;
【基金】：2014年國家電網(wǎng)公司發(fā)展項目“無人機巡檢實用化關鍵技術及檢測體系研究”
【分類號】：TM75;TP391.41
[Abstract]:With the application of UAV (UAV) in power line inspection, the demand of UAV patrol image information mining or target recognition is becoming more and more intense. Traditional power component recognition processes often use classical machine learning algorithms such as support vector machine (SVM) (SVM), random forest or adaboost, combined with gradient color or texture to identify power components. It is difficult to make full use of the image information of UAV patrol, and it is difficult to achieve high accuracy. Convolutional neural network (CNN) is the best algorithm for target recognition because of its excellent performance in target recognition. The region based convolution neural network (RCNN) detects and recognizes the target by using CNN to extract the region that may contain the target from the image, but the computation is complicated. Fast R-CNN and Faster RCNN use CNN network to extract image features, followed by a regional proposal layer, which optimizes the way of extracting possible target areas and improves the classifier for target recognition. The detection and recognition of target is almost in real time. This paper describes the flow of Faster R-CNN algorithm in detail, and uses it in the detection of UAV power line inspection image components, and then compares and analyzes the DPM,SPPnet and Faster R-CNN recognition methods, respectively. The three methods are tested and verified by the data set constructed from the actual data collected from the patrol inspection of power widget, and the influence of different parameters on the identification results is discussed. The experimental results show that the recognition method based on depth learning is feasible. Moreover, the recognition speed of 80 ms and the accuracy of 92.7% can be achieved by using Faster R-CNN to identify and locate various kinds of power components.
【作者單位】： 國網(wǎng)山東省電力公司電力科學研究院國家電網(wǎng)公司電力機器人技術實驗室;山東魯能智能技術有限公司;國網(wǎng)山東省電力公司;
【基金】：2014年國家電網(wǎng)公司發(fā)展項目“無人機巡檢實用化關鍵技術及檢測體系研究”
【分類號】：TM75;TP391.41
【參考文獻】
相關期刊論文 前2條
1 苑津莎;崔克彬;李寶樹;;基于ASIFT算法的絕緣子視頻圖像的識別與定位[J];電測與儀表;2015年07期
2 金立軍;胡娟;閆書佳;;基于圖像的高壓輸電線間隔棒故障診斷方法[J];高電壓技術;2013年05期
相關博士學位論文 前1條
1 吳慶崗;復雜背景輸電線圖像中部件邊緣提取算法研究[D];大連海事大學;2012年
相關碩士學位論文 前1條
1 曹婧;航拍輸電線路圖像中絕緣子部件的提取[D];大連海事大學;2012年
【共引文獻】
相關期刊論文 前10條
1 翟永杰;王迪;趙振兵;程海燕;;基于空域形態(tài)一致性特征的絕緣子串定位方法[J];中國電機工程學報;2017年05期
2 王萬國;田兵;劉越;劉O，
本文編號：2358602
本文鏈接：http://sikaile.net/kejilunwen/ruanjiangongchenglunwen/2358602.html
最近更新
教材專著
