Image detection and segmentation method and system based on deep learning

The embodiment of the invention provides an image detection and segmentation method and system based on deep learning, and belongs to the technical field of image recognition. The method comprises the following steps: acquiring an image to be detected and segmented; obtaining a candidate box in the image based on a grey level histogram distribution method; a CNN network is adopted to classify each candidate frame, so that the category and the corresponding probability of each candidate frame can be obtained; and selecting and segmenting the first N candidate frames as the region where the target is located according to the sequence of the probabilities from large to small. Through the above technical scheme, the image detection and segmentation method and system based on deep learning provided by the invention combine histogram gray calculation with the CNN network, and the histogram gray calculation reduces the overall algorithm complexity and reduces the target detection range of the CNN network at the same