Rotation-based object-oriented ensemble in land use land cover classification of hyperspectral data

Shah, Maqsood Hussain; Liu, Yazhou; Hayat, Hassan

doi:10.1109/intech.2016.7845045

Cited by 2 publications

(1 citation statement)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this sense, pixel-wise classification methods suffer a high intra-class variability (for instance, class-centered pixels and frontier pixels that belong to the same class), and also the high inter-class similarity (frontier pixels that belong to different classes). Furthermore, the so-called "salt and pepper" noise problem appears frequently, since these methods ignore the spatial dependencies between neighboring pixels [44].…”

Section: A Traditional Machine Learning Methods For Spectral-spatialmentioning

confidence: 99%

Rotation Equivariant Convolutional Neural Networks for Hyperspectral Image Classification

et al. 2020

View full text Add to dashboard Cite

Detection of surface material based on hyperspectral imaging (HSI) analysis is an important and challenging task in remote sensing. It is widely known that spectral-spatial data exploitation performs better than traditional spectral pixel-wise procedures. Nowadays, convolutional neural networks (CNNs) have shown to be a powerful deep learning (DL) technique due their strong feature extraction ability. CNNs not only combine spectral-spatial information in a natural way, but have also shown to be able to learn translation-equivariant representations, i.e. a translation of input features into an equivalent internal CNN feature map. This provides great robustness to spatial feature locations. However, as far as we know, CNNs do not exhibit a natural way to exploit rotation equivariance, i.e. make use of the fact that data patches in a HSI data cube are observed in different orientations due to their orientation or on the varying paths/orbits of the airborne/spaceborne spectrometers. This paper presents a rotation-equivariant CNN2D model for HSI analysis, where traditional convolution kernels have been replaced by circular harmonic filters (CHFs). The obtained results over three well-known HSI datasets showcase the potential of the approach. INDEX TERMS Hyperspectral imaging (HSI), convolutional neural network (CNN), harmonic network (H-net), rotation invariance

show abstract

Section: A Traditional Machine Learning Methods For Spectral-spatialmentioning

confidence: 99%