ResNet50模型在肺炎识别分类中的应用OA
Application of ResNet50 in Pneumonia Identification and Classification
为了提高医学图像中肺炎的识别和分类效率,本文使用ResNet50模型对COVID图像、Lung_Opacity图像、Normal图像和Pneumonia图像进行识别分类.通过比较ResNet50 与AlexNet和GoogLeNet对肺炎图像分类的准确率、F1 值、召回率、精确率、特异性评价指标的差别,表明了ResNet50 较其他模型具有更好的图像识别和分类性能.
In order to improve the recognition and classification efficiency of pneumonia in medical images,this paper uses the ResNet50 model to recognize and classify COVID images,Lung-Opacity images,Normal images,and Pneumonia images.By comparing the differences in accuracy,F1 value,recall,accuracy,and specificity evaluation indicators between ResNet50 and AlexNet and GoogLeNet for pneumonia image classification,it is demonstrated that ResNet50 has better image recognition and classification performance than other models.
彭航;邓锡泽;牛玉霞;刘洋
北方民族大学计算机科学与工程学院 银川 750021||北方民族大学图像图形智能处理国家民委重点实验室 银川 750021
临床医学
肺炎图像识别残差网络模型肺炎分类
PneumoniaImage RecognitionResNet50 ModelClassification of Pneumonia
《福建电脑》 2024 (004)
9-13 / 5
本文得到宁夏自然科学基金(No.2023AAC03293)资助.
评论