eduzhai > Applied Sciences > Engineering >

Evolving Deep Convolutional Neural Networks for Hyperspectral Image Denoising

  • king
  • (0) Download
  • 20210507
  • Save

... pages left unread,continue reading

Document pages: 8 pages

Abstract: Hyperspectral images (HSIs) are susceptible to various noise factors leadingto the loss of information, and the noise restricts the subsequent HSIs objectdetection and classification tasks. In recent years, learning-based methodshave demonstrated their superior strengths in denoising the HSIs.Unfortunately, most of the methods are manually designed based on the extensiveexpertise that is not necessarily available to the users interested. In thispaper, we propose a novel algorithm to automatically build an optimalConvolutional Neural Network (CNN) to effectively denoise HSIs. Particularly,the proposed algorithm focuses on the architectures and the initialization ofthe connection weights of the CNN. The experiments of the proposed algorithmhave been well-designed and compared against the state-of-the-art peercompetitors, and the experimental results demonstrate the competitiveperformance of the proposed algorithm in terms of the different evaluationmetrics, visual assessments, and the computational complexity.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×