Infrared small target detection (IRSTD) is the process of separating targets from complex and blurry backgrounds, with wide applications in rescue, maritime guidance, and other fields. However, traditional deep learning methods that rely solely on CNN-based methods are prone to losing targets as the convolution deepens, resulting in missed detections when segmenting small infrared targets. Similarly, because of the unique characteristics of infrared small targets, Transformer-based methods lack the supplement of convolutional local detail features, making it difficult to achieve good results. To address these issues, we propose the Feature Adaptive Transformer and CNN network (FATCNet): a dual branch decoder structure network that utilizes CNN and Transformer to obtain local and global features in parallel. Meanwhile, we propose the Attention Guided Adaptive Fusion Block (AGAF) to better integrate global and local feature information, eliminate semantic gaps, and dynamically fuse features based on network learning. The network encoder incorporates the Hierarchical Attention Module (HAM), which leverages the distinct characteristics of various hierarchical feature maps to enhance features, thereby mitigating the risk of losing crucial details pertaining to infrared small targets. Additionally, We have constructed a bird infrared dataset, named BSIRST v2, which comprises 1000 authentic images with diverse target shapes, sizes, and cluttered backgrounds, and the dataset includes meticulous pixel-level annotations. Our approach's efficacy has been validated through experiments conducted on both public datasets and BSIRST v2. In comparison to other state-of-the-art methods, our FATCNet outperforms in terms of probability of detection (PD), F1 score (F1), and intersection over union (IoU).

Infrared small target detection (IRSTD) has significant value in many civilian and military applications. However, most current methods still struggle to maintain tiny targets with low contrast and are prone to false alarms caused by confusing noise similar to the targets. Furthermore, challenges remain in accurately segmenting target shapes due to the difficulty in acquiring sufficient fine-grained details. To address these issues, we propose the AER-Net to extract discriminative features of small targets while leveraging the distinct semantics of different layers for reducing false alarms and preserving edge contours of targets. Specifically, we present a U-shaped architecture tailored for IRSTD by a Feature Representation Enhancement Module (FREM), which combines the local correlation of CNNs and the global correlation of Transformers to enhance features while preventing the loss of tiny targets. Based on this, we further devise a Daisy-shaped Correlation-aware Module (DCM) to capture sufficient global contexts for mitigating clutter interference in deeper layers. Subsequently, a Gated Refinement Module (GRM) is designed to refine low-level features from the encoder utilizing high-level features as guidance for preserving local details and further removing noise in shallower layers. Moreover, we integrate our method with the monitoring device to extend the application of IRSTD to automatic monitoring of birds in plateau lakes and wetlands, while contributing an infrared bird dataset named BSIRST V1 with targets exhibiting high morphological uncertainty to advance research in this field. Extensive experiments on three public and the proposed datasets have demonstrated that our method achieves a balance between detection capability and segmentation accuracy. Notably, our approach has significant improvements in mean IoU metric on the challenging benchmark NUDT-SIRST and IRSTD-1k datasets with 4.46% and 3.86%, respectively. The source code and dataset are available at https://github.com/fuqingzhang/AER-Net.

In order to obtain a steam-stable hydrogen permselectivity membrane, with tetraethylorthosilicate (TEOS) as the silicon source, zirconium nitrate pentahydrate (Zr(NO₃)₄·5H₂O) as the zirconium source, and methyltriethoxysilane (MTES) as the hydrophobic modifier, the methyl-modified ZrO₂-SiO₂ (ZrO₂-MSiO₂) membranes were prepared via the sol-gel method. The microstructure and gas permeance of the ZrO₂-MSiO₂ membranes were studied. The physical-chemical properties of the membranes were characterized by Fourier transform infrared spectroscopy (FTIR), X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD), transmission electron microscopy (TEM), scanning electron microscope (SEM), and N₂ adsorption–desorption analysis. The hydrogen permselectivity of ZrO₂-MSiO₂ membranes was evaluated with Zr content, temperature, pressure difference, drying control chemical additive (glycerol) content, and hydrothermal stability as the inferred factors. XRD and pore structure analysis revealed that, as n_Zr increased, the MSiO₂ peak gradually shifted to a higher 2θ value, and the intensity gradually decreased. The study found that the permeation mechanism of H₂ and other gases is mainly based on the activation–diffusion mechanism. The separation of H₂ is facilitated by an increase in temperature. The ZrO₂-MSiO₂ membrane with n_Zr = 0.15 has a better pore structure and a suitable ratio of micropores to mesopores, which improved the gas permselectivities. At 200 °C, the H₂ permeance of MSiO₂ and ZrO₂-MSiO₂ membranes was 3.66 × 10⁻⁶ and 6.46 × 10⁻⁶ mol·m⁻²·s⁻¹·Pa⁻¹, respectively. Compared with the MSiO₂ membrane, the H₂/CO₂ and H₂/N₂ permselectivities of the ZrO₂-MSiO₂ membrane were improved by 79.18% and 26.75%, respectively. The added amount of glycerol as the drying control chemical additive increased from 20% to 30%, the permeance of H₂ decreased by 11.55%, and the permselectivities of H₂/CO₂ and H₂/N₂ rose by 2.14% and 0.28%, respectively. The final results demonstrate that the ZrO₂-MSiO₂ membrane possesses excellent hydrothermal stability and regeneration capability.

Methyl-modified, cobalt-doped silica (Co/MSiO₂) materials were synthesized by sol-gel technique calcined in N₂ atmospheres, and membranes were made thereof by coating method. The effects of Co/Si molar ratio (n_Co) on the physical-chemical constructions of Co/MSiO₂ materials and microstructures of Co/MSiO₂ membranes were systematically investigated. The gas permeance performance and hydrothermal stability of Co/MSiO₂ membranes were also tested. The results show that the cobalt element in Co/MSiO₂ material calcined at 400 °C exists not only as Si–O–Co bond but also as Co₃O₄ and CoO crystals. The introduction of metallic cobalt and methyl can enlarge the total pore volume and average pore size of the SiO₂ membrane. The activation energy (E_a) values of H₂, CO₂, and N₂ for Co/MSiO₂ membranes are less than those for MSiO₂ membranes. When operating at a pressure difference of 0.2 MPa and 200 °C compared with MSiO₂ membrane, the permeances of H₂, CO₂, and N₂ for Co/MSiO₂ membrane with n_Co = 0.08 increased by 1.17, 0.70, and 0.83 times, respectively, and the perm-selectivities of H₂/CO₂ and H₂/N₂ increased by 27.66% and 18.53%, respectively. After being steamed and thermally regenerated, the change of H₂ permeance and H₂ perm-selectivities for Co/MSiO₂ membrane is much smaller than those for MSiO₂ membrane.

Modifying PVDF membranes with Al₂O₃ nanoparticles via two methods provided a new idea for removing dyes. It also elucidatend the retention mechanism of the modified membranes when treating different dye solutions.