测绘学报
測繪學報
측회학보
ACTA GEODAETICA ET CARTOGRAPHICA SINICA
2015年
8期
909-918
,共10页
张春森%郑艺惟%黄小兵%崔卫红
張春森%鄭藝惟%黃小兵%崔衛紅
장춘삼%정예유%황소병%최위홍
光谱-空间特征%概率融合%支持向量机%高光谱%分类
光譜-空間特徵%概率融閤%支持嚮量機%高光譜%分類
광보-공간특정%개솔융합%지지향량궤%고광보%분류
spectral-spatial feature%probabilistic fusion%SVM%hyperspectral%classification
提出了一种基于光谱-空间多特征加权概率融合的高光谱影像分类方法.首先,利用最小噪声分离(minimum noise fraction,MNF)方法对高光谱影像进行降维和特征提取,并以得到的MNF 特征影像作为光谱特征,联合灰度共生矩阵(gray level co-occurrence matrix,GLCM)提取的纹理特征、基于OFC算子建立的多尺度形态学特征以及采用连续最大角凸锥(sequential maximum angle convex cone,SMACC)提取的端元组分特征,组成3组光谱-空间特征;然后利用支持向量机(support vector machine,SVM)对每一组光谱-空间特征进行分类,得到每组特征的概率输出结果;最后,建立多特征加权概率融合模型,应用该模型将不同特征的概率输出结果进行加权融合,得到最终分类结果.为了验证该方法的有效性,利用 ROSIS和 AVIRIS影像进行试验,总体分类精度分别达到97.65%和96.62%.结果表明本文的方法不但较好地克服了传统基于单一特征高光谱影像分类的局限性,而且其分类效果也优于常规矢量叠加(vector stacking,VS)和概率融合的多特征分类方法,有效地改善了高光谱影像的分类结果.
提齣瞭一種基于光譜-空間多特徵加權概率融閤的高光譜影像分類方法.首先,利用最小譟聲分離(minimum noise fraction,MNF)方法對高光譜影像進行降維和特徵提取,併以得到的MNF 特徵影像作為光譜特徵,聯閤灰度共生矩陣(gray level co-occurrence matrix,GLCM)提取的紋理特徵、基于OFC算子建立的多呎度形態學特徵以及採用連續最大角凸錐(sequential maximum angle convex cone,SMACC)提取的耑元組分特徵,組成3組光譜-空間特徵;然後利用支持嚮量機(support vector machine,SVM)對每一組光譜-空間特徵進行分類,得到每組特徵的概率輸齣結果;最後,建立多特徵加權概率融閤模型,應用該模型將不同特徵的概率輸齣結果進行加權融閤,得到最終分類結果.為瞭驗證該方法的有效性,利用 ROSIS和 AVIRIS影像進行試驗,總體分類精度分彆達到97.65%和96.62%.結果錶明本文的方法不但較好地剋服瞭傳統基于單一特徵高光譜影像分類的跼限性,而且其分類效果也優于常規矢量疊加(vector stacking,VS)和概率融閤的多特徵分類方法,有效地改善瞭高光譜影像的分類結果.
제출료일충기우광보-공간다특정가권개솔융합적고광보영상분류방법.수선,이용최소조성분리(minimum noise fraction,MNF)방법대고광보영상진행강유화특정제취,병이득도적MNF 특정영상작위광보특정,연합회도공생구진(gray level co-occurrence matrix,GLCM)제취적문리특정、기우OFC산자건립적다척도형태학특정이급채용련속최대각철추(sequential maximum angle convex cone,SMACC)제취적단원조분특정,조성3조광보-공간특정;연후이용지지향량궤(support vector machine,SVM)대매일조광보-공간특정진행분류,득도매조특정적개솔수출결과;최후,건립다특정가권개솔융합모형,응용해모형장불동특정적개솔수출결과진행가권융합,득도최종분류결과.위료험증해방법적유효성,이용 ROSIS화 AVIRIS영상진행시험,총체분류정도분별체도97.65%화96.62%.결과표명본문적방법불단교호지극복료전통기우단일특정고광보영상분류적국한성,이차기분류효과야우우상규시량첩가(vector stacking,VS)화개솔융합적다특정분류방법,유효지개선료고광보영상적분류결과.
A hyperspectral images classification method based on the weighted probabilistic fusion of multiple spectral-spatial features was proposed in this paper.First,the minimum noise fraction (MNF)approach was employed to reduce the dimension of hyperspectral image and extract the spectral feature from the image,then combined the spectral feature with the texture feature extracted based on gray level co-occurrence matrix (GLCM),the multi-scale morphological feature extracted based on OFC operator and the end member feature extracted based on sequential maximum angle convex cone (SMACC)method to form three spectral-spatial features.Afterwards,support vector machine (SVM)classifier was used for the classification of each spectral-spatial feature separately.Finally, we established the weighted probabilistic fusion model and applied the model to fuse the SVMoutputs for the final classification result. In order to verify the proposed method,the ROSIS and AVIRIS image were used in our experiment and the overall accuracy reached 97.65% and 96.62% separately.The results indicate that the proposed method can not only overcome the limitations of traditional single-feature based hyperspectral image classification,but also be superior to conventional VS-SVM method and probabilistic fusion method.The classification accuracy of hyperspectral images was improved effectively.