摘要
针对尿沉渣中的有形成分进行检测和分析,提出结合主成分分析(PCA)和极限学习机(ELM)的识别和统计方法。该方法通过PCA对样本进行特征提取和降维后输入到ELM进行训练,根据训练得出的模型与未经PCA处理的样本训练的模型进行检测效果对比。实验结果表明,使用PCA处理后的样本训练得出的模型具有更高的识别准确度和稳定性,同时训练时间大幅减少。
The recognition and statistics method combining principal component analysis(PCA)and extreme learning machine(ELM)is proposed to detect and analyze the visible components in urine sediment.The features of the sample are extracted by means of PCA and then input to extreme learning machine(ELM)for training after dimensionality reduction.The detection effects of the trained model got by training and sample training model without PCA processing are compared.The experimental result shows that the model obtained by sample training after PCA processing has higher recognition accuracy and stability,and its training time is greatly reduced.
作者
秦传波
冯宝
谌瑶
QIN Chuanbo;FENG Bao;CHEN Yao(College of Information Engineering,Wuyi University,Jiangmen 529020,China;School of Electronic Information and Automation,Guilin University of Aerospace Technology,Guilin 541004,China)
出处
《现代电子技术》
北大核心
2019年第11期45-49,共5页
Modern Electronics Technique
基金
国家自然科学基金(61771347)
青年创新人才类项目(2015KQNCX165)
五邑大学青年科研基金(2015zk10)
江门市科技计划项目(江科[2017]268)
大学生创新创业训练计划项目(201611349025,201711349086,201711349024)~~
关键词
尿沉渣检测
尿沉渣成分分类
极限学习机
主成分分析
图像识别
特征提取
医学显微图像
urine sediment detection
urine sediment component classification
extreme learning machine
principal component analysis
image recognition
feature extraction
medical microscopic image
