Big data technology is more and more widely used in modern power systems. Efficient collection of big data such as equipment status, maintenance and grid operation in power systems, and data mining are the important research topics for big data application in smart grid. In this paper, the application of big data technology in fast image recognition of transmission towers which are obtained using fixed-wing unmanned aerial vehicle (UAV) by large range tilt photography are researched. A method that using fast region-based convolutional neural networks (Rcnn) convolutional architecture for fast feature embedding (Caffe) to get deep learning of the massive transmission tower image, extract the image characteristics of the tower, train the tower model, and quickly recognize transmission tower image to generate power lines is proposed. The case study shows that this method can be used in tree barrier modeling of transmission lines, which can replace artificial identification of transmission tower, to reduce the time required for tower identification and generating power line, and improve the efficiency of tree barrier modeling by around 14.2%.