Automatic bridge crack detection system fusing cyclic residual convolution and context extractor network

The invention relates to the technical field of image recognition, in particular to an automatic bridge crack detection system fusing cyclic residual convolution and a context extractor network. The method comprises the following steps: acquiring a bridge crack image by using image acquisition equipment, and creating a bridge crack data set for deep learning model training; creating a novel feature encoder-decoder network in which the standard convolution in the encoder is replaced by a cyclic residual convolution block (RRCNN); using a context extractor network comprising cavity convolution, a dense cavity convolution block (DAC) and a residual multi-core pooling block (RMP); constructing a bridge crack automatic detection model by combining a novel feature encoder-decoder network and a context extractor network; training the bridge crack automatic detection model through the bridge crack data set, and obtaining ideal accuracy; and according to the parameters obtained by training, inputting a to-be-detected