Method and system for detecting nipple positions in molybdenum target image

The embodiment of the invention provides a method and system for detecting the nipple position in a molybdenum target image. The method comprises the steps of inputting a to-be-detected molybdenum target image into a trained neural network model, and obtaining an output result of the trained neural network model; determining a nipple position in the molybdenum target image to be detected accordingto the output result; wherein the trained neural network model is obtained by training according to a training set, and the training set comprises a plurality of sample molybdenum target images and asample label set corresponding to each sample molybdenum target image. According to the method and the system provided by the embodiment of the invention, through a deep learning technology, the sameneural network model can be used for carrying out nipple position detection on different types of molybdenum target images, the robustness and universality of an algorithm are improved, and the accuracy of a detection result i