Recursive autoencoders based unsupervised feature learning for hyperspectral image classification

      Research output: Research - peer-reviewArticle

      Early online date
      • Xiangrong Zhang
      • Yanjie Liang
      • Chen Li
      • Licheng Jiao
      • Huiyu Zhou

      View graph of relations

      For hyperspectral image (HSI) classification, it is very important to learn effective features for the discrimination purpose. Meanwhile, the ability to combine spectral and spatial information together in a deep level is also important for feature learning. In this letter, we propose an unsupervised feature learning method for HSI classification, which is based on recursive autoencoders (RAE) network. RAE utilizes the spatial and spectral information and produces high-level features from the original data. It learns features from the neighborhood of the investigated pixel to represent the whole local homogeneous area of the image. In addition, to obtain more accurate representation of the investigated pixel, a weighting scheme is adopted based on the neighboring pixels, where the weights are determined by the spectral similarity between the neighboring pixels and the investigated pixel. The effectiveness of our method is evaluated by the experiments on two hyperspectral datasets and the results show that our proposed method has better performance.

      Documents

      DOI

      Original languageEnglish
      Number of pages5
      JournalIEEE Geoscience and Remote Sensing Letters
      Journal publication date11 Oct 2017
      Early online date11 Oct 2017
      DOIs
      StateEarly online date - 11 Oct 2017

      ID: 132277188