The complex environment along railway lines, characterized by low imaging quality, strong background interference, and densely distributed small objects, causes existing detection models to suffer from low accuracy in practical applications. To tackle these challenges, this study aims to develop a robust and lightweight insulator detection model specifically optimized for these challenging railway scenarios. To this end, we release a dedicated comprehensive dataset named complexRailway that covers typical railway scenarios to address the limitations of existing insulator datasets, such as the lack of small-scale objects in high-interference backgrounds. On this basis, we present CutP5-LargeKernelAttention-SIoU (C5LS), an improved YOLOv8 variant with three key improvements: (1) optimized YOLOv8’s detection head by removing the P5 branch to improve feature extraction for small- and medium-sized targets while reducing computational redundancy, (2) integrating a lightweight Large Separable Kernel Attention (LSKA) module to expand the receptive field and improve contextual modeling, (3) and replacing CIoU with SIoU loss to refine localization accuracy and accelerate convergence. Experimental results demonstrate that it reaches 94.7% in mAP@0.5 and 65.5% in mAP@0.5–0.95, outperforming the baseline model by 1.9% and 3.5%, respectively. With an inference speed of 104 FPS and a model size of 13.9 MB, the model balances high precision and lightweight deployment. By providing stable and accurate insulator detection, C5LS not only offers reliable spatial positioning basis for subsequent defect identification but also builds an efficient and feasible intelligent monitoring solution for these failure-prone insulators, thereby effectively enhancing the operational safety and maintenance efficiency of the railway power system.

The environmental impact of trade liberalization hinges on a central tension between its ‘scale effect’ and its ‘technique and composition effects’. This study uses the Regional Comprehensive Economic Partnership’s (RCEP) influence on China-ASEAN green agricultural trade as a quasi-natural experiment to test the net outcome of this conflict. Leveraging panel data from 2003–2023 and a combined Difference-in-Differences (DID) and Instrumental Variables (IV) methodology, we find that while RCEP-driven agricultural expansion (the scale effect) did significantly increase NO₂ emissions by 21.80 units (p = 0.011), the total effect of RCEP was counter-intuitively and significantly negative. The agreement produced a context-specific environmental co-benefit, achieving a net reduction of 99.45 $\mathsf{\mu }\mathrm{m}\mathrm{o}\mathrm{l}/{\mathrm{m}}^2$ of NO₂ (p = 0.008). The agreement produced a specific reduction in combustion-related air pollution, achieving a net reduction of 99.45 $\mathsf{\mu }\mathrm{m}\mathrm{o}\mathrm{l}/{\mathrm{m}}^2$ of NO₂ (p = 0.008). We caution that this outcome is contextual: it is driven not only by a still-weak green technology effect (a 0.55 unit reduction, p = 0.095) but more critically by a structural shift in trade towards lower-emission fruit and vegetable products (coefficient = −0.13, p = 0.077), suggesting that environmental gains are currently contingent on product composition rather than broad decarbonization. This reveals that environmental pressures from scale are, for now, being offset by structural optimization, highlighting the urgent need for refined emission monitoring and targeted green policies within RCEP to solidify these environmental gains.