山东大学学报(工学版)
山東大學學報(工學版)
산동대학학보(공학판)
JOURNAL OF SHANDONG UNIVERSITY(ENGINEERING SCIENCE)
2015年
1期
37-44,63
,共9页
方向梯度直方图特征%滑动窗口%非最大值抑制%组织病理图像%细胞检测
方嚮梯度直方圖特徵%滑動窗口%非最大值抑製%組織病理圖像%細胞檢測
방향제도직방도특정%활동창구%비최대치억제%조직병리도상%세포검측
HOG feature%sliding window%non-maxima suppression%histopathological image%nuclei detection
提出一种基于方向梯度直方图(histograms of oriented gradient,HOG)特征和滑动窗口的细胞检测方法,能快速、高效、准确地检测高分辨率病理组织图像中的细胞。该检测算法首先对训练集中的细胞样本块和非细胞样本块提取HOG特征,然后运用HOG特征训练分类器。训练好的分类器用于在整幅病理图像中自动检测细胞。先运用滑动窗的方法在整幅高分辨率病理图像中选取相同尺寸的所有可能的细胞块,被滑动窗选定的图像块提取HOG特征后,送到训练好的分类器中判断是否是细胞块。为了验证提出方法的有效性,将此方法运用于17名乳腺患者的共37张H&E(hematoxylin&eosin)染色高分辨率穿刺切片病理图像上自动检测细胞,通过与softmax (SM)分类器、稀疏自编码器+SM、局部二值模式+SM、支持向量机(support vector machine,SVM)、HOG+SVM、以及 HOG+SVM多个模型对细胞检测的准确率、召回率以及综合评价指标的对比表明,本研究提出的方法分别为71.5%,82.3%和76.5%,具有更高的准确率。
提齣一種基于方嚮梯度直方圖(histograms of oriented gradient,HOG)特徵和滑動窗口的細胞檢測方法,能快速、高效、準確地檢測高分辨率病理組織圖像中的細胞。該檢測算法首先對訓練集中的細胞樣本塊和非細胞樣本塊提取HOG特徵,然後運用HOG特徵訓練分類器。訓練好的分類器用于在整幅病理圖像中自動檢測細胞。先運用滑動窗的方法在整幅高分辨率病理圖像中選取相同呎吋的所有可能的細胞塊,被滑動窗選定的圖像塊提取HOG特徵後,送到訓練好的分類器中判斷是否是細胞塊。為瞭驗證提齣方法的有效性,將此方法運用于17名乳腺患者的共37張H&E(hematoxylin&eosin)染色高分辨率穿刺切片病理圖像上自動檢測細胞,通過與softmax (SM)分類器、稀疏自編碼器+SM、跼部二值模式+SM、支持嚮量機(support vector machine,SVM)、HOG+SVM、以及 HOG+SVM多箇模型對細胞檢測的準確率、召迴率以及綜閤評價指標的對比錶明,本研究提齣的方法分彆為71.5%,82.3%和76.5%,具有更高的準確率。
제출일충기우방향제도직방도(histograms of oriented gradient,HOG)특정화활동창구적세포검측방법,능쾌속、고효、준학지검측고분변솔병리조직도상중적세포。해검측산법수선대훈련집중적세포양본괴화비세포양본괴제취HOG특정,연후운용HOG특정훈련분류기。훈련호적분류기용우재정폭병리도상중자동검측세포。선운용활동창적방법재정폭고분변솔병리도상중선취상동척촌적소유가능적세포괴,피활동창선정적도상괴제취HOG특정후,송도훈련호적분류기중판단시부시세포괴。위료험증제출방법적유효성,장차방법운용우17명유선환자적공37장H&E(hematoxylin&eosin)염색고분변솔천자절편병리도상상자동검측세포,통과여softmax (SM)분류기、희소자편마기+SM、국부이치모식+SM、지지향량궤(support vector machine,SVM)、HOG+SVM、이급 HOG+SVM다개모형대세포검측적준학솔、소회솔이급종합평개지표적대비표명,본연구제출적방법분별위71.5%,82.3%화76.5%,구유경고적준학솔。
A new method was presented which integrated histograms of oriented gradient (HOG)feature and sliding window for rapid,efficient and accurate detection of nuclei from high resolution pathological images.HOG feature was extracted from the training samples which include both nuclei and non-nuclei patches.The supervised classifier were trained with HOG features.The trained classifier was employed for automated nuclei detection from input patches that selected from histopathological images.During the detection,sliding window was used to select patches.In order to verify the effectiveness of the method on detecting nuclei from histopathological images,this article compared the pro-posed method with softmax (SM)classifier,sparse autoencoder(SAE)+SM,local binary pattern (LBP)+SM,sup-port vector machine(SVM),HOG+SM,and HOG+SVM models.The experiments on 37 pieces of H&E staining his-topathological images showed that the proposed method achieved highest precision,recall and F1-measure values,which were 7 1.5%,82.3% and 76.5% respectively.