Insulator target detection method and system based on transfer learning

The invention discloses an insulator target detection method and system based on transfer learning. The method comprises the following steps: constructing a target detection network; performing iterative training on the target detection network by adopting a preset knowledge distillation strategy according to an output result of a preset teacher network, and updating learning parameters in the knowledge distillation strategy according to an efficiency jump learning strategy network in each iterative training process to obtain a final insulator detection model; the ultraviolet image containing the insulator is input into the insulator detection model, the insulator detection model outputs a segmented image corresponding to the ultraviolet image, and the segmented image only contains the insulator. 本发明公开了一种基于迁移学习的绝缘子目标检测方法及系统，方法包括：构建目标检测网络；根据预设的教师网络的输出结果，采用预设的知识蒸馏策略对目标检测网络进行迭代训练，并在每次迭代训练过程中，根据效能跃升学习策略网络对知识蒸馏策略中的学习参数进行更新，得到最终的绝缘子检测模型；将包含绝缘子的紫外图像输入至绝缘子检测模型中，绝缘子检测模型输出与紫外图像相对应的分割图像，分割图像中仅包含绝缘子。