一種基于SVM的無人機(jī)影像中單個(gè)建筑物的角點(diǎn)檢測方法
發(fā)布時(shí)間：2018-07-16 17:31
【摘要】：針對(duì)目前無人機(jī)影像中單個(gè)建筑物角點(diǎn)的檢測現(xiàn)狀,提出了一種基于支持向量機(jī)(SVM)的無人機(jī)影像中建筑物的角點(diǎn)檢測方法。首先對(duì)4個(gè)波段的無人機(jī)影像進(jìn)行多尺度分割,計(jì)算影像的NDVI,通過植被與非植被區(qū)域的波譜差異剔除植被的影響。其次,用面向?qū)ο蠓诸惙▽?建筑物塊"從影像中提取出來,對(duì)"建筑物塊"區(qū)域用Harris算子進(jìn)行邊緣檢測,形成建筑物邊緣點(diǎn)集數(shù)據(jù)。隨后通過設(shè)計(jì)高斯徑向基將邊緣樣本點(diǎn)映射到高維特征空間,構(gòu)建特征向量,采用邊緣點(diǎn)集訓(xùn)練SVM分類模型,最終通過SVM分類模型從粗提取的邊緣點(diǎn)集中檢測出正確的建筑物角點(diǎn),實(shí)現(xiàn)了單個(gè)建筑物的角點(diǎn)提取。
[Abstract]:According to the current situation of single building corner detection in UAV image, a new method of building corner detection in UAV image based on support vector machine (SVM) is proposed. Firstly, the multi-scale segmentation of UAV images in four bands was carried out, the NDVI of the images was calculated, and the influence of vegetation was eliminated by the spectral differences between vegetation and non-vegetation regions. Secondly, the "building block" is extracted from the image by the object-oriented classification method, and the edge of the "building block" area is detected by Harris operator to form the building edge point set data. Then, the edge sample points are mapped to the high dimensional feature space by designing Gao Si radial basis function, and the feature vectors are constructed, and the classification model is trained by edge point set. Finally, the correct corner points of buildings are detected from rough edge points by SVM classification model, and the corner points of a single building are extracted.
【作者單位】： 桂林理工大學(xué)測繪地理信息學(xué)院;廣西空間信息與測繪重點(diǎn)實(shí)驗(yàn)室;南寧市勘察測繪地理信息院;
【基金】：國家自然科學(xué)基金(41161073) 廣西自然科學(xué)基金(2016GXNSFAA380013;2014GXNSFDA118038) 桂林市科學(xué)研究與技術(shù)開發(fā)計(jì)劃(2016012601) 重慶基礎(chǔ)科學(xué)與前沿技術(shù)研究項(xiàng)目(cstc2015jcyj B028)
【分類號(hào)】：P237
本文編號(hào)：2127107
[Abstract]:According to the current situation of single building corner detection in UAV image, a new method of building corner detection in UAV image based on support vector machine (SVM) is proposed. Firstly, the multi-scale segmentation of UAV images in four bands was carried out, the NDVI of the images was calculated, and the influence of vegetation was eliminated by the spectral differences between vegetation and non-vegetation regions. Secondly, the "building block" is extracted from the image by the object-oriented classification method, and the edge of the "building block" area is detected by Harris operator to form the building edge point set data. Then, the edge sample points are mapped to the high dimensional feature space by designing Gao Si radial basis function, and the feature vectors are constructed, and the classification model is trained by edge point set. Finally, the correct corner points of buildings are detected from rough edge points by SVM classification model, and the corner points of a single building are extracted.
【作者單位】： 桂林理工大學(xué)測繪地理信息學(xué)院;廣西空間信息與測繪重點(diǎn)實(shí)驗(yàn)室;南寧市勘察測繪地理信息院;
【基金】：國家自然科學(xué)基金(41161073) 廣西自然科學(xué)基金(2016GXNSFAA380013;2014GXNSFDA118038) 桂林市科學(xué)研究與技術(shù)開發(fā)計(jì)劃(2016012601) 重慶基礎(chǔ)科學(xué)與前沿技術(shù)研究項(xiàng)目(cstc2015jcyj B028)
【分類號(hào)】：P237
【相似文獻(xiàn)】
相關(guān)期刊論文 前1條
1 閆利,林宗堅(jiān),于子凡;基于結(jié)構(gòu)分析的線劃角點(diǎn)檢測[J];武漢測繪科技大學(xué)學(xué)報(bào);1996年03期
相關(guān)會(huì)議論文 前1條
1 張睿;陳雪;馬建文;;基于遞歸SVM的高光譜數(shù)據(jù)特征選擇算法研究[A];第八屆成像光譜技術(shù)與應(yīng)用研討會(huì)暨交叉學(xué)科論壇文集[C];2010年
相關(guān)碩士學(xué)位論文 前1條
1 張漢女;基于SVM的海岸線提取方法研究[D];東北師范大學(xué);2010年
，本文編號(hào)：2127107
本文鏈接：http://www.sikaile.net/kejilunwen/dizhicehuilunwen/2127107.html
最近更新
教材專著
