In order to accurately identify the digital operators of distribution network and improve the effect of remote infrared monitoring of digital operation safety of distribution network, a remote infrared monitoring system of digital operation safety of distribution network based on laser nonlinear imaging technology is studied. The system entails the design of a comprehensive digital image acquisition process covering the SF6 layer and the infrared application layer. Furthermore, it encompasses the hardware configuration of a remote infrared monitoring terminal tailored for the digital operation safety of distribution networks, aligned with the structural and imaging principles of SF6 laser nonlinear cameras. The software component orchestrates the remote infrared monitoring process, incorporating a convolutional neural network algorithm to construct a multi scale pyramid for operator feature extraction. Pixel features are then enhanced through fast local Laplace filter transformation. Intrusion behavior detection within hazardous areas is achieved via softmax function, facilitating safe remote infrared monitoring of distribution network digital operations. The experimental results show that the image enhancement processing ability and operator recognition effect of the system are ideal, and can timely and effectively warn the intrusion behavior of operators in the dangerous area. The warning time is always less than 0.06 s, and the comprehensive performance of network transmission is excellent. The maximum amount of data packet loss is 5 000, which improves the effect of safe remote infrared monitoring of digital operation in distribution network.