Abstract To effectively mitigate weed threats to corn seedlings, improve recognition accuracy, and meet the requirements for mobile deployment, this study proposed a lightweight model named YOLOv8n-DSSW, which was developed based on YOLOv8n. Several key improvements were introduced: the C2f_Dual module was integrated to achieve initial lightweighting; the SPDConv module was employed to enhance the detection capability for small targets and low-resolution images while further reducing model comp