Implementation and analysis of a lightweight node.js map recognition model
To realize"problem map"recognition,using the training of network model,from massive network images quickly identify the map images,through the Google company proposed MobileNet network recognition model transformation,transfer learning,realize a small convolution neural network model,quickly identify network map images thin image recognition model.The results indicate that models can greatly accelerate work efficiency,it is one of the more convenient and fast methods in map review work.
万方数据
