Automatic Recognition and Classification Algorithm of Medical Images Based on Neural Network Machine Learning

At present, image recognition technology is in the development era. Target recognition has become one of the most important research contents in the field of deep learning. How to identify the target accurately and effectively becomes the key, especially the convolution model with excellent performance is used as the model of target detection, which has very critical and important research value in the field of target detection and identification. From the perspective of several kinds of network models and referring to other models, this thesis makes structural improvement and comparison, showing its good structural advantages and effectively solving the problems existing in some target detection. The experimental results show that the network model mentioned in this thesis can be well used for image information recognition and other items, especially for face recognition and other detection tasks. Compared with the original convolution network, the model has higher accuracy, effectively reduces the computation of network parameters, reduces the consumption of hardware memory resources, and has good stability.