Infrared dim and small target detection is one of the crucial technologies in the military field, but it faces various challenges such as weak features and small target scales. To overcome these challenges, this paper proposes IR-TransDet, which integrates the benefits of the convolutional neural network and the Transformer to properly extract global semantic information and features of small targets. First, the Efficient Feature Extract Module (EFEM) is designed, which uses depthwise convolution and pointwise convolution to effectively capture the features of the target. Then, an improved Residual Sim Atrous Spatial Pyramid Pooling (ASPP) module is proposed based on the image characteristics of infrared dim and small targets. The proposed method focuses on enhancing the edge information of the target. Meanwhile, an IR-Transformer module is devised, which uses the self-attention mechanism to investigate the relationship between the global image, the target, and neighboring pixels. Finally, experiments were conducted on four open datasets, and the results indicate that IR-TransDet achieves state-of-the-art performance in infrared dim and small target detection. To achieve a comparative evaluation of the existing infrared dim and small target detection methods, this study constructed the ISTD-Benchmark tool, which is available at https://linaom1214.github.io/ISTD-Benchmark.

Infrared multi-band small and dim target detection is an important research direction in the fields of modern remote sensing and military surveillance. However, achieving high-precision detection remains challenging due to the small scale, low contrast of small and dim targets, and their susceptibility to complex background interference. This paper innovatively proposes a dual-band infrared small and dim target detection method (MM-IRSTD). In this framework, we integrate a convolutional self-attention mechanism module and a self-distillation mechanism to achieve end-to-end dual-band infrared small and dim target detection. The Conv-Based Self-Attention module consists of a convolutional self-attention mechanism and a multilayer perceptron, effectively extracting and integrating input features, thereby enhancing the performance and expressive capability of the model. Additionally, this module incorporates a dynamic weight mechanism to achieve adaptive feature fusion, significantly reducing computational complexity and enhancing the model’s global perception capability. During model training, we use a spatial and channel similarity self-distillation mechanism to drive model updates, addressing the similarity discrepancy between long-wave and mid-wave image features extracted through deep learning, thus improving the model’s performance and generalization capability. Furthermore, to better learn and detect edge features in images, this paper designs an edge extraction method based on Sobel. Finally, comparative experiments and ablation studies validate the advancement and effectiveness of our proposed method.

Dual-band infrared small-target detection offers richer information than single-band methods but also poses two key challenges: modeling weak targets in cluttered scenes and robustly fusing the complementary features of the two bands. To overcome these challenges, we propose the state space model and self-attention for infrared dual-band small target detection (s3-irdstd) framework. This framework first employs a novel dynamic attention module that integrates state-space modeling with self-attention, efficiently capturing long-range dependencies and global context to enhance target discriminability. Second, to ensure effective feature integration, we design a Fourier-based fusion module. Operating in the frequency domain, it avoids the information loss that can occur in direct spatial fusion, thereby yielding a more complete target representation. Experiments show that s³-irdstd significantly outperforms existing approaches, validating its effectiveness and innovative design. Further information is available on our project page: https://linaom1214.github.io/SSD.

Traditional metallic interdigital electrodes are rigid and undeformable, flexible interdigital capacitors are therefore appealing as strain sensors. In this study, interdigital capacitors were parametrically designed by 3D printing and encapsulated by spraying process. The interdigital circuits of the structure were printed with conductive silicone rubber (CSR) filled with silver-coated glass fiber and carbon fiber (CF), and the circuits were encapsulated with polydimethylsiloxane (PDMS). Herein, the interdigital-flexible structures were parametrically designed and firstly served as capacitive sensor, namely flexible interdigital capacitive sensors (FICS). The spaces between interdigital electrodes (IEs), are extremely sensitive to strain, therefore provide the capacitors with excellent electromechanical behaviors. The optimized FICS benefited for a wide working range of strain (0~45%), high sensitivity (gauge factor, GF=2.7) to a tiny strain of 0.3 %, stable working duration at different stretching speeds (18 mm/min, 36 mm/min and 72 mm/min), prolonged service life (>800 cycles), as well as excellent capability to detect human movement (bulging, grasping and bending). Response mechanism of the FICS was modelled based upon its microstructure evolution, including the distances between IEs and the fillers migration. The printed FICS with optimized structure provide a comprehensive thought in the design of electronics, further would inspire the branch of 3d printed electronics.