Continuous self-learning image recognition method and system

The invention relates to the technical field of continuous self-learning image recognition, in particular to a continuous self-learning image recognition method and system. The system comprises a continuous self-learning image recognition image server. The continuous self-learning image recognition image server comprises an image acquisition module, a picture primary marking module, an image recognition model training module, a new picture marking module, an automatic/manual picture verification module and an optimal model generation module. According to the invention, the model trained by theinitial training picture set is automatically utilized to predict the new picture, the classification result of the new picture is automatically or manually verified, finally, the verified picture isused as the new training set and combined with the old training set to be used for model training again, the model is iteratively trained, and the continuous optimization of models is realized. 本发明涉及持续自学习的图像识别技术领域，尤其为一种持续自学习的