Scholarlink

In this chapter we present a detailed introduction to deep learning and more specifically its use in data fusion. In the first half of the chapter we concentrate on the convolutional neural network (CNN) discussing: training the CNN, CNN architectures (VGGNet, GoogleNet, ResNet, DenseNet, SiameseNet), recent developments (Image fusion CNN) and hybrid networks (SIFT-CNN). In the second half of the chapter we turn to unsupervised deep learning concentrating on autoencoders (AE) discussing AE architectures (convolutional, denoising, denseBlock) and recent developments (denseFuse). Altogether the chapter includes more than 15 examples illustrating CNN’s and AE’s in different computer vision applications.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Deep Learning

  • 章节

  • 发表日期：

    2026年

查看全文
  • Harvey B. Mitchell

摘要

In this chapter we present a detailed introduction to deep learning and more specifically its use in data fusion. In the first half of the chapter we concentrate on the convolutional neural network (CNN) discussing: training the CNN, CNN architectures (VGGNet, GoogleNet, ResNet, DenseNet, SiameseNet), recent developments (Image fusion CNN) and hybrid networks (SIFT-CNN). In the second half of the chapter we turn to unsupervised deep learning concentrating on autoencoders (AE) discussing AE architectures (convolutional, denoising, denseBlock) and recent developments (denseFuse). Altogether the chapter includes more than 15 examples illustrating CNN’s and AE’s in different computer vision applications.