【计算机科学】【2017】深度学习方法及其应用——交通标志分类与老年痴呆症的图像检测

本文为瑞典查尔姆斯理工大学（作者：LINNéA CLAESSON、BJÖRN HANSSON）的硕士论文，共94页。

本文将深度学习方法（卷积神经网络CNN）用于解决交通标志识别和老年痴呆症检测这两种分类问题。使用的两个数据集来自德国交通标志识别基准（GTSRB）和老年痴呆症神经成像倡议（ADNI）。对交通标志数据集的最终测试结果产生了98.81%的分类准确率，几乎达到人类对同一数据集的识别准确率98.84%。还测试了所选CNN结构的不同参数设置，以了解它们对分类精度的影响。试图区分健康大脑和患有老年痴呆症大脑的MRI图像仅获得了约65%的分类准确率。这些结果表明，卷积神经网络方法在交通标志的分类中是非常有前途的，但是在处理更复杂的老年痴呆症检测问题时还需要进行更多的工作。

In this thesis, the deep learning methodconvolutional neural networks (CNNs) has been used in an attempt to solve twoclassification problems, namely traffic sign recognition and Alzheimer’sdisease detection. The two datasets used are from the German Traffic SignRecognition Benchmark (GTSRB) and the Alzheimer’s Disease NeuroimagingInitiative (ADNI). The final test results on the traffic sign dataset generateda classification accuracy of 98.81 %, almost as high as human performance onthe same dataset, 98.84 %. Different parameter settings of the selected CNNstructure have also been tested in order to see their impact on the classificationaccuracy. Trying to distinguish between MRI images of healthy brains and brainsafflicted with Alzheimer’s disease gained only about 65 % classification accuracy.These results show that the convolutional neural network approach is very promisingfor classifying traffic signs, but more work needs to be done when working withthe more complex problem of detecting Alzheimer’s disease.

下载英文原文地址：

http://page2.dfpan.com/fs/9lcjc2211291662fb68/

更多精彩文章请关注微信号：