Cross-modal remote sensing object detection holds significant potential for around-the-clock applications. However, the modality differences between cross-modal data and the degradation of feature quality under adverse weather conditions limit detection performance. To address these challenges, this paper presents a novel cross-modal remote sensing object detection framework designed to overcome two critical challenges in around-the-clock applications: (1) significant modality disparities between visible light, infrared, and synthetic aperture radar data, and (2) severe feature degradation under adverse weather conditions including fog, and nighttime scenarios. Our primary contributions are as follows: First, we develop a multi-scale feature extraction module that employs a hierarchical convolutional architecture to capture both fine-grained details and contextual information, effectively compensating for missing or blurred features in degraded visible-light images. Second, we introduce an innovative feature interaction module that utilizes cross-attention mechanisms to establish long-range dependencies across modalities while dynamically suppressing noise interference through adaptive feature selection. Third, we propose a feature correction fusion module that performs spatial alignment of object boundaries and channel-wise optimization of global feature consistency, enabling robust fusion of complementary information from different modalities. The proposed framework is validated on visible light, infrared, and SAR modalities. Extensive experiments on three challenging datasets (LLVIP, OGSOD, and Drone Vehicle) demonstrate our framework’s superior performance, achieving state-of-the-art mean average precision scores of 66.3%, 58.6%, and 71.7%, respectively, representing significant improvements over existing methods in scenarios with modality differences or extreme weather conditions. The proposed solution not only advances the technical frontier of cross-modal object detection but also provides practical value for mission-critical applications such as 24/7 surveillance systems, military reconnaissance, and emergency response operations where reliable around-the-clock detection is essential.

The automatic segmentation of the heart from echocardiography can significantly improve the efficiency and effectiveness of assessing cardiac function and provide objective and reproducible quantitative evidence for clinical diagnosis, treatment planning, and disease screening. Many excellent results in the community of medical artificial intelligence have been achieved; however, there is still room for improvement, such as the employment of temporal information in echocardiography. In this paper, we propose the Spiking U-Mamba model, which incorporates three novel modules, i.e., MultiStepLIFEncoder, BidirectionalFeatureFusion and FeatureRefinementBlock. MultiStepLIFEncoder encodes static features into spatio-temporal pulse sequences to enhance the model’s boundary resolution capability, while BidirectionalFeatureFusion enables bidirectional interaction and attention enhancement between the Mamba global features and pulse features, and FeatureRefinementBlock further refines the fused features and ensures stable gradient propagation. Two public echocardiography datasets, CAMUS and EchoNet-Dynamic, were employed to evaluate the proposed Spiking U-Mamba model. On the CAMUS dataset, Spiking U-Mamba achieved a Dice coefficient of 90.16% during the ED phase and 90.54% during the ES phase, outperforming other SOTA methods; on the EchoNet-Dynamic dataset, Spiking U-Mamba also performed well. Since the employment of the Spiking Neural Network, the proposed Spiking U-Mamba model achieved a good balance between segmentation accuracy and computational cost.