Further results on dissimilarity spaces for hyperspectral images RF-CBIR
This work addresses the challenge of enhancing CBIR systems for hyperspectral images, which is an incremental improvement for remote sensing and image analysis domains.
The paper tackled the problem of applying content-based image retrieval (CBIR) with relevance feedback (RF) to hyperspectral images by proposing a dissimilarity spaces approach, which improved performance over previous methods based on spectral/spatial features or dictionaries, as validated on a real dataset.
Content-Based Image Retrieval (CBIR) systems are powerful search tools in image databases that have been little applied to hyperspectral images. Relevance feedback (RF) is an iterative process that uses machine learning techniques and user's feedback to improve the CBIR systems performance. We pursued to expand previous research in hyperspectral CBIR systems built on dissimilarity functions defined either on spectral and spatial features extracted by spectral unmixing techniques, or on dictionaries extracted by dictionary-based compressors. These dissimilarity functions were not suitable for direct application in common machine learning techniques. We propose to use a RF general approach based on dissimilarity spaces which is more appropriate for the application of machine learning algorithms to the hyperspectral RF-CBIR. We validate the proposed RF method for hyperspectral CBIR systems over a real hyperspectral dataset.