Convolutional neural networks (CNNs) have recently found great attention in image classification since deep CNNs have exhibited excellent performance in computer vision. Owing to their immense success, of late, scientists are exploring the functionality of transformers in Earth observation applications. Nevertheless, the primary issue with transformers is that they demand significantly more training data than CNN classifiers. Thus, the use of these transformers in remote sensing is considered challenging, notably in utilizing polarimetric SAR (PolSAR) data, due to the insufficient number of existing labeled data. In this letter, we develop and propose a vision transformer-based framework that utilizes 3D and 2D CNNs as feature extractors and, in addition, local window attention for the effective classification of PolSAR data. Extensive experimental results demonstrated that the developed model PolSARFormer obtained better classification accuracy than the state-of-the-art vision Swin Transformer and FNet algorithms. The PolSARFormer outperformed the Swin Transformer and FNet by the margin of 5.86% and 17.63%, in terms of average accuracy in the San Francisco data benchmark. Moreover, the results over the Flevoland dataset illustrated that the PolSARFormer exceeds several other algorithms, including the ResNet (97.49%), Swin Transformer (96.54%), FNet (95.28%), 2D CNN (94.57%), and AlexNet (91.83%), with a kappa index of 99.30%. The code will be made available publicly at https://github.com/aj1365/PolSARFormer.

Accurate spatio-temporal information about rice growth is an important factor for agronomic management and regional grain yield estimation. In this letter, a unified framework for monitoring and mapping of rice using dense time-series of Sentinel-1 synthetic aperture radar (SAR) images is proposed. A processing chain for such dense time-series Sentinel-1 images is developed with the Google Earth Engine's cloud computing platform. A dense time-series analysis of backscatter response of rice with different management practices is analyzed. Subsequently, the early and late transplanted rice is classified using a clustering algorithm within this platform. The proposed approach is used to monitor different cultivars of rice in three districts in the state of West Bengal, which is one of the major rice growing regions in India. The classification accuracy is assessed across 150 validation points spanning multiple blocks for the 2017 monsoon season. The Sentinel-1 SAR images acquired up to the early vegetative stage for rice have provided satisfactory classification accuracy with an overall accuracy >85% with κ ~ 0.86 across different management practices throughout the region.

The classification of urban areas in polarimetric synthetic aperture radar (PolSAR) data is a challenging task. Moreover, urban structures oriented away from the radar line of sight pose an additional complexity in the classification process. The characterization of such areas is important for disaster relief and urban sprawl monitoring applications. In this paper, a novel technique based on deep learning is proposed, which leverages a synthetic target database for data augmentation. The PolSAR dataset is rotated by uniform steps and collated to form a reference database. A stacked autoencoder network is used to transform the information in the augmented dataset into a compact representation. This significantly improves the generalization capabilities of the network. Finally, the classification is performed by a multilayer perceptron network. The modular architecture allows for easy optimization of the hyperparameters. The synthetic target database is created and the classification performance is evaluated on an L-band airborne UAVSAR dataset and L-band space-borne ALOS-2 dataset acquired over San Francisco, USA. The proposed technique shows an overall accuracy of $91.3{\%}$. An improvement over state-of-the-art techniques is achieved, especially in urban areas rotated away from the radar line of sight.

In radar polarimetry, incoherent target decomposition techniques help extract scattering information from polarimetric synthetic aperture radar (SAR) data. This is achieved either by fitting appropriate scattering models or by optimizing the received wave intensity through the diagonalization of the coherency (or covariance) matrix. As such, the received wave information depends on the received antenna configuration. Additionally, a polarimetric descriptor that is independent of the received antenna configuration might provide additional information which is missed by the individual elements of the coherency matrix. This implies that existing target characterization techniques might neglect this information. In this regard, we suitably utilize the 2-D and 3-D Barakat degree of polarization which is independent of the received antenna configuration to obtain distinct polarimetric information for target characterization. In this study, we introduce new roll-invariant scattering-type parameters for both full-polarimetric (FP) and compact-polarimetric (CP) SAR data. These new parameters jointly use the information of the 2-D and 3-D Barakat degree of polarization and the elements of the coherency (or covariance) matrix. We use these new scattering-type parameters, which provide equivalent information as the Cloude α for FP SAR data and the ellipticity parameter χ for CP SAR data, to characterize various targets adequately. Additionally, we appropriately utilize these new scattering-type parameters to obtain unique non-model-based three-component scattering power decomposition techniques. We obtain the even-bounce, and the odd-bounce scattering powers by modulating the total polarized power by a proper geometrical factor derived using the new scattering-type parameters for FP and CP SAR data. The diffused scattering power is obtained as the depolarized fraction of the total power. Moreover, due to the nature of its formulation, the decomposition scattering powers are non-negative and roll-invariant while the total power is conserved. The proposed method is both qualitatively and quantitatively assessed utilizing the L-band ALOS-2 and C-band Radarsat-2 FP and the associated simulated CP SAR data.

Crop growth monitoring using compact-pol synthetic aperture radar (CP-SAR) data is gaining attention with the rapid advancements toward operational applications. In this article, we propose a vegetation index for compact polarimetric (CP) SAR data [compact-pol radar vegetation index (CpRVI)]. The CpRVI is derived using the concept of a geodesic distance between the Kennaugh matrices projected on a unit sphere. This distance is utilized to compute a similarity measure between the observed Kennaugh matrix and the Kennaugh matrix of an ideal depolarizer (a realization of vegetation canopy). The similarity measure is then modulated with a scaled quantity derived from the scattering power ratio of the same and opposite sense polarization with respect to the transmitted circular polarization. In this article, we utilize time-series-simulated RADARSAT Constellation Mission (RCM) compact-pol SAR data (RH-RV) obtained from the full-pol RADARSAT-2 observations during the soil moisture active passive (SMAP) validation experiment 2016 (SMAPVEX16-MB) campaign in Manitoba, Canada, to assess the proposed vegetation index. Among the various crops grown in this region, in particular, we analyze the growth stages of wheat and soybean due to their different canopy structures. A temporal analysis of the proposed CpRVI with crop biophysical parameters [the plant area index (PAI) and vegetation water content (VWC)] at different phenological stages confirms the trend of CpRVI with the plant growth. Nevertheless, variations of CpRVI values are apparent with different plant densities for both the crop types. Also, the linear regression analysis confirms that the CpRVI values significantly correlate with PAI (r = 0.72 and 0.85) and VWC (r = 0.62 and 0.75) for both wheat and soybean. We observed good retrieval of PAI and VWC for both wheat and soybean.

Shallow convolutional neural networks (CNNs) have successfully been used to classify polarimetric synthetic aperture radar (PolSAR) imagery. However, one drawback of the existing deep CNN-based techniques is that the input PolSAR training data are often insufficient due to their need for a significant number of training data compared to shallow CNN models utilized in PolSAR image classification. In this letter, we propose using the Haar wavelet transform in deep CNNs for effective feature extraction to improve the classification accuracy of PolSAR imagery. Based on the results, the proposed deep CNN model obtained better average accuracy in the San Francisco region with an accuracy of 93.3% and produced more homogeneous classification maps with less noise compared to the two much shallower CNN models of AlexNet (87.8%) and a 2-D CNN network (91%). The proposed algorithm is efficient and may be applied over large areas to support regional wetland mapping and monitoring activities using PolSAR imagery. The codes are available at ( https://github.com/aj1365/DeepCNN_Polsar ).