A bone age evaluation method based on two-stage neural network

The invention provides a bone age evaluation method based on a two-stage neural network, which comprises the following steps: 1. Adjusting the original input hand bone image size to 512x512, namely scaling according to scale, and adding edge information; 2, extract an image mask; 3, generate a training image; 4, respectively calculating two-dimensional single-scale wavelet transform and Canny operator edge extraction for the image generated in the step 3, and composing three channels of input samples together with the image equalized by the histogram; 5, carrying out data enhancement on the three-channel image generated in the step 4 to expand the data set size; 6. Prediction of bone age; By combining the image segmentation and bone age prediction, the detection method can suppress the noise and segment the hand bone area accurately, so as to evaluate the bone age. For other image processing fields such as target segmentation, target retrieval, target regression prediction and so on, it has practical applicat