Aspect-based sentiment analysis aims to predict sentiment polarity for every aspect in a sentence review. Most existing approaches are based on the sequence models, which may superimpose the emotional semantics of different tendencies and lack syntactic structure information. And most models adopt coarse-grained attention mechanism which still face the issues of weakness interaction between aspect and context. In this paper, we propose a transformer based multi-grained attention network (T-MGAN), which utilizes the Transformer module to learn the word-level representations of aspects and context respectively, and further utilizes the Tree Transformer module to obtain the phrase-level representations of contexts. It is capable of extracting the syntactic structure features and syntax information of aspect and context. In addition, we adopt dual-pooling method and multi-grained attention network to extract high quality aspect-context interactive representations. We evaluate the proposed model on three datasets and prove the effectiveness of the proposed model.

Due to the unique imaging mechanism of synthetic aperture radar (SAR), which leads to a discrete state of aircraft targets in images, its detection performance is vulnerable to the influence of complex ground objects. Although existing deep learning detection algorithms show good performance, they generally use a feature pyramid neck design and large backbone network, which reduces the detection efficiency to some extent. To address these problems, we propose a simple and efficient attention network (SEAN) in this paper, which takes YOLOv5s as the baseline. First, we shallow the depth of the backbone network and introduce a structural re-parameterization technique to increase the feature extraction capability of the backbone. Second, the neck architecture is designed by using a residual dilated module (RDM), a low-level semantic enhancement module (LSEM), and a localization attention module (LAM), substantially reducing the number of parameters and computation of the network. The results on the Gaofen-3 aircraft target dataset show that this method achieves 97.7% AP at a speed of 83.3 FPS on a Tesla M60, exceeding YOLOv5s by 1.3% AP and 8.7 FPS with 40.51% of the parameters and 86.25% of the FLOPs.

This paper proposes an information enhancement network based on self-supervised learning (SEL-Net) for runway area detection. During the self-supervised learning phase, the distinctive attributes of PolSAR multi-channel data are fully harnessed to enhance the generated pretrained model’s focus on airport runway areas. During the detection phase, this paper presents an improved U-Net detection network. Edge Feature Extraction Modules (EEM) are integrated into the encoder and skip connection sections, while Semantic Information Transmission Modules (STM) are embedded into the decoder section. Furthermore, improvements have been applied to the network’s upsampling and downsampling architectures. Experimental results demonstrate that the proposed SEL-Net effectively addresses the issues of high false alarms and runway integrity, achieving a superior detection performance.

In this paper, a new algorithm is proposed for civil airport runway detection in polarimetric synthetic aperture radar (POLSAR) image based on unsupervised classification. First, the contents of POLSAR image are classified into ten categories with scattering characteristics of terrain and Wishart classifier; then suspected airport areas are extracted from classification result based on the power property of terrain; Structural features are used for final detection of the true runway. Experimental results with measured POLSAR data demonstrate the validity of the proposed method which has a high correct detection rate and a low false alarm rate.

Land Use and Land Cover (LULC) classification is one of the tasks of Polarimetric Synthetic Aperture Radar (PolSAR) images’ interpretation, and the classification performance of existing algorithms is highly sensitive to the class number, which is inconsistent with the reality that LULC classification should have multiple levels of detail in the same image. Therefore, an object-oriented unsupervised classification algorithm for PolSAR images based on the image block is proposed. Firstly, the image is divided into multiple non-overlapping image blocks, and $h/q/gray$-Wishart classification is performed in each block. Secondly, each cluster obtained is regarded as an object, and the affinity matrix of objects is calculated in the global image. Finally, the objects are merged into the specified class number by density peak clustering (DPC), and the adjacent objects at the block boundary are checked and forced to merge. Experiments are carried out with the measured data of the airborne AIRSAR and E-SAR and the spaceborne GF-3. The experimental results show that the proposed algorithm achieves good classification results under a variety of class numbers.

The gain flatness is crucial to characterize bandwidth and measurement precision for the probes in the near-field measurements. In this communication, an electric-field probe (E-field probe) with enhanced gain flatness is proposed, fabricated, and calibrated. It consists of the coupling section made up of circle-shaped tip to enhance the gain flatness in the lower frequency, the transition section to suppress the resonances or the ripples in the middle and higher frequency, and the transmission section with well-matched impedance cascaded to the instrument. A symmetric chamfered bottom edge (SCBE) is optimized to suppress the unwanted energy divergence propagating from the coupling section to the transmission section. The field distribution is analyzed to aid the optimal design for the probe. To further have a deep insight into the enhanced gain flatness, quasi-FFT (QFFT) method and the feature selective validation (FSV) method are applied in the simulation as well as the measurement. The measured results demonstrate a maximum fluctuation of approximately 2.1 dB and an average fluctuation of 1.45 dB. Comparisons with other referenced probes are also conducted, confirming the enhanced gain flatness of the proposed E-field probe.