Construction site fence area positioning method and system based on deep learning

The invention discloses a construction site fence area positioning method and system based on deep learning, and the method comprises the steps: 1) extracting a red fence in a construction site picture according to the color features of the fence, and then converting the red fence into a binary image to train an ENet network, thereby obtaining a trained ENet network model; 2) utilizing the trained ENet network model to segment fences in the construction site picture to obtain red fences in the fences; and 3) according to the segmented red fence, carrying out rectangular filling and polygonal expansion on pixels of the red fence to generate a surface area covering all the fence, and meanwhile, marking a position corresponding to the position of the surface area in the construction site picture to realize the positioning of the construction site fence area. The method can quickly and accurately segment the position of the fence on the construction site, and has the advantages of low cost, high accuracy, low cal