The work aims to propose a method of detecting improper operations in logistics with lightweight and enhanced countermeasures in view of the difficulty in effectively identifying improper operations in logistics under the background of complex and strong noise. Based on YOLOv5, the lightweight GhostC3 module was offered, and the lightweight countermeasure module was proposed with the idea of countermeasure learning. The C3 module in the original structure was changed into the lightweight GhostC3 module, the Conv module was changed into the LAconv module, and the positioning loss was changed into CIOU loss. Finally, through experimental verification, the method proposed had an excellent detection effect against improper operations under the background of complex and strong noise. Compared with YOLOv5, the average detection accuracy of the method proposed increased by 1.69%, the number of model parameters decreased by 45.14%, and the detection speed was improved by 2.46%. The method proposed has the characteristics of a low number of parameters, fast detection speed, and high accuracy. It is advanced and practical for the detection of improper operations in logistics under the background of complex and strong noise, and fully meets the detection needs of improper operations in logistics.