【目的】 随着国际贸易的频繁开展，口岸有害生物检疫防控工作的重要性与日俱增。天牛作为常见且具有潜在危害的昆虫，实现其在口岸环境下的准确识别，对于防范外来物种入侵、保障生态安全意义重大。本研究旨在开发一款高效、便捷的口岸天牛识别小程序，为口岸检疫工作提供技术支持。【方法】 本研究设计并开发口岸天牛识别小程序，其核心包含基于YOLOv5s的天牛检测模型与基于改进ResNet50的天牛识别模型。一方面，利用YOLOv5s模型在目标检测领域高效快速的优势，针对天牛图像特征进行参数优化，实现天牛在图像中的精准定位；另一方面，对ResNet50识别模型进行改进，融入卷积块注意力模块（Convolutional block attention module，CBAM）注意力机制，深度挖掘天牛外观关键特征，突破相似种类识别难点。最后，将两个模型集成至小程序中，并设计简洁操作界面，方便口岸工作人员实时上传天牛图像进行检测与识别。【结果】 经大量测试样本验证及实际口岸场景应用显示，该小程序能够在复杂环境下，快速且准确地检测和识别天牛种类。无论是面对不同光照条件、背景干扰，还是天牛处于不同生长阶段、姿态的图像，小程序均能稳定输出可靠结果。【结论】 本研究开发的口岸天牛识别小程序显著提升了口岸天牛检疫工作效率，为防范外来天牛入侵筑牢技术防线，为生态安全保障提供了有力支持。该研究成果对推动口岸有害生物智能识别技术的发展具有积极的促进作用，具备良好的实际应用价值与推广潜力。

  [Aim]  To develop an efficient and convenient mini-program for the identification of Cerambycidae species at ports, thereby providing useful technical support for port quarantine work. [Methods]  This study designed and developed a mini-program for identifying members of the Cerambycidae at ports, with its core consisting of Cerambycidae detection models based on YOLOv5s and the improved ResNet50. The mini-program leverages the efficient and rapid advantages of the YOLOv5s model in object detection, allowing parameter optimization to be carried out based on image characteristics of the Cerambycidae to achieve precise identification of members of the Cerambycidae in images. The ResNet50 identification model was improved by integrating the CBAM (Convolutional Block Attention Module) to extract key features of the Cerambycidae and thereby overcome difficulties in identifying similar species. The integration of both models into the mini-program, together with a simple user interface, allows port quarantine staff to upload Cerambycidae images in real-time for detection and identification. [Results]  Verification with a large number of test samples in actual port scenarios indicates that the mini-program quickly, and accurately, detects and identifies members of the Cerambycidae in complex environments, and produces reliable results under different lighting conditions and in the presence of background interference, irrespective of the growth stage and posture of the insect. [Conclusion]  A mini-program developed for identifying members of the Cerambycidae at ports significantly improves the efficiency of port quarantine work, thereby helping prevent the establishment of alien Cerambycidae. This new program promotes the development of intelligent identification technology for harmful organisms at ports, and has good practical application value and potential for promotion.