Infrared small target detection is one of the vital tasks in various infrared detection application, and has some typical challenges such as small and dim target, background noise, and complex scenes. To address the problem, a context guided reverse attention network (CgraNet) is proposed to detect infrared small target by introducing context guided module (CGM), multiscale aggregation block (MAB), and reverse attention module (RAM). The CGM is designed to capture inherent property of semantic information from multi-scale encode layer in pixel-level recognition. In order to eliminate the impact of low-level feature on computational complexity and ensure the detection performance, we design the MAB to aggregate multiscale feature. The RAM is integrated in decoder layer to combine the features from MAB and CGM for fusing the localization information and multi-scale structural information. Extensive experiments on infrared small target datasets demonstrate that our method can achieve high detection accuracy and low false alarm rate compared with some state-of-the-art model-driven and data-driven methods.
A chaotic adaptive seeker optimization algorithm (CASOA) is proposed in this study to improve the coupling efficiency and accuracy of a butterfly optical communication laser. It primarily relies on chaotic disturbance to improve seeker search performance. The chaotic disturbance enables the algorithm to jump out from local extremes. Furthermore, chaos is associated with a novel strategy for optimizing search paths with a small population. A simulation and experiment are conducted to demonstrate that the CASOA with a few seekers has an excellent search success rate with few iterations in the coupling alignment. These results indicate that the proposed CASOA can reliably improve the coupling accuracy and efficiency of laser diodes and single-mode fibers.
Improving the spot brightness and uniformity of arrangement of the array laser is conducive to ensuring the beam quality of the fiber laser. Based on the light tracing principle, the optical model performance of two common fiber lasers was first analyzed. Then, a novel rotationally polarized optical model with high power and spot uniformity was designed and optimized on the basis of the aforementioned analysis. The results of the evaluation metrics of the multi-indicator optical model show that the spot uniformity of our proposed model improved by 24.03%, the power improved by 0.55%, and the maximum light distance was shortened from 120 mm to 82.58 mm. Further, the results of the coupling tolerance analysis of the optical elements show that the total coupling efficiency was 89.04%. The coupling power and tolerance relationships did not produce degradation compared with the traditional model. Extensive comparative results show that the designed rotationally polarized optical path model can effectively improve the optical coupling efficiency and spot uniformity of arrayed semiconductor lasers.
11 June 2025 in Measurement Science and Technology
With the development of image recognition technology, small-target detection has attracted wide interest in early warning systems due to the fact that it operates in real-time and has a high accuracy rate. To further improve the accuracy and robustness of small-target recognition in the field of infrared surveillance, based on the mechanism of global feature capture and local detail enhancement, we propose a local-patch and global attention network (LPGANet) with multiscale asymmetric fusion based on binary segmentation. This allows us to efficiently capture local-patch feature details and global position information of small targets for segmentation between targets and backgrounds. In particular, the multiscale asymmetry fusion module utilizes a combination of deep and shallow semantic features to achieve pixel preservation at key up-sampled locations. By fusing multiscale feature maps at different levels, the target detection accuracy can be effectively improved. The microscopic recognition module (MRM) can effectively perform feature extraction for target weak edges, and by introducing MRM into the global network, which can optimize the extraction ability of dim and dark target edges, and ensure the robustness of network detection. Extensive experimental results show that LPGANet is effective in recognizing small infrared targets from the background and that it demonstrates high accuracy and stable detection results.
1 January 2026 in IEEE Transactions on Industrial Electronics