Single-frame infrared small and dim target detection underpins military early warning and space surveillance. However, existing review literature lacks a classification system based on architectural paradigms and standardized cross-architecture benchmarking. This paper systematically reviews deep learning methods from two complementary perspectives: classification based on architectural paradigms and quantitative empirical evaluation. Within our proposed analytical framework, we categorize existing methods into seven architectural paradigms—detector-based, U-Net-based, GAN-based, Transformer-based, diffusion-based, Mamba-based, and hybrid models. Each paradigm is defined by two inseparable components: the “task formulation” and the “computational architecture.” The task formulation specifies how the detection problem is transformed into a mathematical problem (e.g., bounding box regression, pixel-level segmentation, conditional denoising generation, etc.), while the computational architecture specifies which computational units are used to implement this formulation (e.g., CNN convolutions, self-attention mechanisms, state space models, etc.). For each paradigm, we analyze the potential conflicts between these paradigms and the physical characteristics of infrared weak targets, as well as the corresponding optimization pathways. Building upon this foundation, through a comprehensive survey of current algorithms, we have selected over 20 representative methods and established a cross-architecture performance comparison benchmark using five metrics (IoU, nIoU, Pd, Fa, and Params) across three datasets (NUAA-SIRST, NUDT-SIRST, and IRSTD-1k). Through three types of diagnostic experiments—performance benchmarking, evaluation of generalization capability across datasets, and robustness testing against synthetic interference sources—this study reveals cross-paradigm patterns, suggesting that within the settings and datasets examined, the primary performance bottleneck appears to have shifted from architectural complexity to the preservation of spatial information during the encoding stage and the learning of semantic discriminative representations. This review can serve as a valuable reference for both newcomers seeking to understand current trends and researchers exploring future directions for infrared weak target detection.

The infrared spectral features of aero-engine wake serve as a crucial basis for infrared detection, tracking, and identification of aircraft. Their spectral line structure is regulated by the concentration and temperature of the exhaust gases, as well as atmospheric absorption effects. This study employs a high-precision numerical simulation model and conducts field experiments on infrared spectral detection of exhaust plumes to systematically investigate the influence mechanisms of three factors—fuel–air ratio (achieved by varying exhaust gas concentration), exhaust temperature, and atmospheric absorption effects—on key spectral features, including peak intensity, line broadening, shape, and position. Results indicate the following. The fuel–air ratio enhances peak intensity and broadens spectral line width by increasing CO₂ and H₂O concentrations within the fuel-lean operating regime (f = 0.025–0.031); exhaust temperature causes exponential spectral intensity growth and line broadening by elevating molecular energy level populations, while inducing H₂O spectral band redshifts and fine-structure evolution; selective absorption by atmospheric H₂O, CO₂, and O₃ causes severe spectral attenuation near bands at 1041 cm⁻¹, 1590 cm⁻¹, 2349 cm⁻¹, and 3756 cm⁻¹, with transmittance distribution modulated by geographical and seasonal factors. This study provides theoretical support for wake spectral feature extraction, infrared stealth design, and detection/tracking identification.