Airplane detection is a challenging problem in complex remote sensing imaging. In this letter, an effective airplane detection framework called Markov random field-fully convolutional network (M-FCN) is proposed. The M-FCN uses a cascade strategy that consists of an FCN-based coarse candidate extraction stage, a multi-Markov random field (multi-MRF)-based region proposal (RP) generation stage, and a final classification stage. In the first stage, the FCN model is trained to be sensitive to airplanes, and a coarse candidate map is generated. This model is scale-, direction-, and color-invariant and does not require many training examples. After the first stage, the coarse candidate map is used as the initial labeling field for a multi-MRF algorithm, and RPs are generated according to the multi-MRF output. This RP-generating strategy can yield more accurate locations with fewer RPs. In the last stage, a convolutional neural network-based classifier is used to improve the precision of the entire framework. Experiments show that the M-FCN has high precision, recall, and location accuracy.

Scene classification has become an active research area in remote sensing (RS) image interpretation. Recently, Transformer-based methods have shown great potential in modeling global semantic information and have been exploited in RS scene classification. In this letter, we propose a multi-level fusion Swin Transformer (MFST), which integrates a multi-level feature merging (MFM) module and an adaptive feature compression (AFC) module to further boost the performance for RS scene classification. The MFM module narrows the semantic gaps in multi-level features via patch merging in lower-level feature maps and lateral connections in the top-down pathway. The AFC module makes multi-level features have smaller dimensions and more coherent semantic information by adaptive channel reduction. We evaluate the proposed network on the aerial image dataset (AID) and NWPU-RESISC45 (NWPU) datasets, and the classification results reveal that the proposed network outperforms several state-of-the-art (SOTA) methods.

To realize real-time imaging for the spaceborne synthetic aperture radar (SAR) using the chirp scaling (CS) algorithm, high generation rates of phase factors and heavy computation loads are unavoidable. For example, updating phase factors continuously in the 2-D time or frequency space is difficult for generating phase factors for real-time imaging. To solve this problem, a modified CS algorithm based on updating phase factors regionally is proposed. Moreover, conventional floating-point arithmetic is unaffordable for real-time imaging onboard. Therefore, fixed-point processing is adopted. In the proposed imaging algorithm, the phase factors are updated only at some specified times and frequencies, and fixed-point operation is adopted. Furthermore, based on the paired echo theory and the finite word length error mathematical model, the effect of the proposed imaging algorithm on the SAR image quality is studied. Moreover, based on the above analysis and the experiments of Chinese HJ-1C and GF-3 spaceborne SAR practical measured data, a hardware system including digital signal processor and field-programmable gate array boards is constructed.

In remote sensing scene classification (RSSC), restrictions on real-time processing on power consumption, performance, and resources necessitate the compression of neural networks. Unlike artificial neural networks (ANNs), spiking neural networks (SNNs) convey information through spikes, offering superior energy efficiency and biological plausibility. However, the high latency of SNNs restricts their practical application in RSSC. Therefore, there is an urgent need to research ultra-low-latency SNNs. As latency decreases, the performance of the SNN significantly deteriorates. To address this challenge, we propose a novel spatio-temporal pruning method that enhances the feature capture capability of ultra-low-latency SNNs. Our approach integrates spatial fundamental structures during the training process, which are subsequently pruned. We conduct a comprehensive evaluation of the impacts of these structures across classic network architectures, such as VGG and ResNet, demonstrating the generalizability of our method. Furthermore, we develop an ultra-low-latency training framework for SNNs to validate the effectiveness of our approach. In this paper, we successfully achieve high-performance ultra-low-latency SNNs with a single time step for the first time in RSSC. Remarkably, our SNN with one time step achieves at least 200 times faster inference time while maintaining a performance comparable to those of other state-of-the-art methods.

Deep learning (DL) has been introduced to change detection (CD) due to its powerful feature representation and robust generalization abilities. However, the application of large DL models has high computational complexity and massive storage requirements for achieving good performance. For disaster emergency response and other applications with high timeliness requirements, it is difficult to deploy large DL models on spaceborne edge devices with limited resources to achieve on-board CD processing. To address this limitation, a novel CD based on knowledge distillation (CDKD) method that combines prototypical contrastive (PC) distillation and channel-spatial-normalized (CSN) distillation is proposed. PC distillation represents the feature distribution by calculating the differences between the similarities of pixel features and their positive and negative prototypes, and improves the student model's detection ability in changed regions that have similar features to the background by mimicking the relative feature distribution. CSN distillation combines two distillation paradigms, channel normalization and spatial normalization, and guides the student model to comprehensively learn the knowledge contained in the output probabilities of the teacher model to accurately identify changed regions with complex shapes. The effectiveness and reliability of the proposed CDKD method are verified on three public remote sensing CD datasets, and extensive experiments and analyses show that the proposed CDKD method can be used to train lightweight models with comparable performance to that of large models.

With the development of satellite load technology and very large scale integrated (VLSI) circuit technology, onboard real-time synthetic aperture radar (SAR) imaging systems have become a solution for allowing rapid response to disasters. A key goal of the onboard SAR imaging system design is to achieve high real-time processing performance with severe size, weight, and power consumption constraints. In this paper, we analyse the computational burden of the commonly used chirp scaling (CS) SAR imaging algorithm. To reduce the system hardware cost, we propose a partial fixed-point processing scheme. The fast Fourier transform (FFT), which is the most computation-sensitive operation in the CS algorithm, is processed with fixed-point, while other operations are processed with single precision floating-point. With the proposed fixed-point processing error propagation model, the fixed-point processing word length is determined. The fidelity and accuracy relative to conventional ground-based software processors is verified by evaluating both the point target imaging quality and the actual scene imaging quality. As a proof of concept, a field- programmable gate array−application-specific integrated circuit (FPGA-ASIC) hybrid heterogeneous parallel accelerating architecture is designed and realized. The customized fixed-point FFT is implemented using the 130 nm complementary metal oxide semiconductor (CMOS) technology as a co-processor of the Xilinx xc6vlx760t FPGA. A single processing board requires 12 s and consumes 21 W to focus a 50-km swath width, 5-m resolution stripmap SAR raw data with a granularity of 16,384 × 16,384.