An Efficient Multi-Sensor Remote Sensing Image Clustering in Urban Areas via Boosted Convolutional Autoencoder (BCAE)

High-resolution urban image clustering has remained a challenging task. This is mainly because its performance strongly depends on the discrimination power of features. Recently, several studies focused unsupervised learning methods by autoencoders to learn and extract more efficient features for purposes. paper proposes Boosted Convolutional AutoEncoder (BCAE) method based feature clustering. The proposed was applied multi-sensor remote-sensing images through multistep workflow. optical data were first preprocessed applying Minimum Noise Fraction (MNF) transformation. Then, these MNF features, in addition normalized Digital Surface Model (nDSM) vegetation indexes such as Normalized Difference Vegetation Index (NDVI) Excess Green (ExG(2)), used inputs BCAE model. Next, our convolutional autoencoder trained automatically encode upgraded boost hand-crafted producing clustering-friendly ones. we employed Mini Batch K-Means algorithm cluster deep Finally, comparative sets manually designed three modes prove efficiency extracting compelling Experiments datasets show learning. According experimental results, method, ultimate become suitable clustering, spatial correlation among pixels process also considered.