摘要
随着机器学习的兴起,基于深度学习的裂缝检测分类得到重视。本文设计了一种基于卷积神经网络结合支持向量机的图像裂缝分类算法,解决了现阶段存在的小样本学习能力不足、分类精度低等问题,实现了对桥梁裂缝图像的有效分类。
With the development of machine learning,crack detection and classification based on deep learning has been paid more attention.This paper designs an image crack classification algorithm based on convolution neural network and support vector machine,which solves the problems of insufficient learning ability of small samples and low classification accuracy,and realizes the effective classification of bridge crack images.
作者
杨济瑞
张晓燕
杜小甫
Yang Jirui;Zhang Xiaoyan;Du Xiaofu(School of Information Science and Technology,Tan Kah Kee College,Xiamen University,Zhangzhou Fujian,363105)
出处
《电子测试》
2021年第4期44-45,104,共3页
Electronic Test
基金
漳州市自然科学基金(ZZ2020J04)资助
厦门大学嘉庚学院预研项目(YY2019L01)
厦门大学嘉庚学院大学生创新创业训练计划项目(202013469020)。
关键词
机器学习
卷积神经网络
支持向量机
裂缝检测
machine learning
convolution neural network
support vector machine
bridge crack
