Domain adaptive model training method and device, image detection method and device, equipment and medium

Abstract

The invention relates to the field of classification models of artificial intelligence, and provides a domain adaptive model training method and device, an image detection method and device, equipmentand a medium. The method comprises the steps: acquiring an image sample set comprising a plurality of image samples, and inputting the image samples into a Faster-RCNN-based domain adaptive model containing the initial parameters; performing image conversion on the image sample through a preprocessing model to obtain a preprocessed image sample; obtaining a feature vector graph through a featureextraction model; obtaining a region feature vector graph through a region extraction model; obtaining a local feature alignment loss value through the local feature model; performing regularization and global feature recognition processing through a global feature model to obtain a feature regularization loss value and a global feature alignment loss value; obtaining a detection loss value through the detection model; obtaining a total loss value; and iteratively updating the initial parameters until convergence to obtain a trained domain adaptive model. According to the invention, the cross-domain image recognition is realized, and the accuracy and reliability of image recognition are improved.

Description

technical field [0001] The invention relates to the field of artificial intelligence classification models, in particular to a domain adaptive model training, image detection method, device, computer equipment and storage medium. Background technique [0002] At present, deep learning methods have been widely used in artificial intelligence. However, deep learning methods are very dependent on the distribution of training data. If there is a difference in the distribution of the collected training data, the detection accuracy of the model finally trained by the deep learning method will be low. For example, OCT (Optical coherence tomography) lesion detection is a very important part of medical diagnosis. Researchers have begun to detect lesions through OCT based on deep learning. However, due to the differences in the acquisition parameters and acquisition methods of different OCT acquisition devices, the distribution of data collected by different devices is different, whi...

AI Technical Summary

Problems solved by technology

Researchers have begun to detect lesions through OCT based on deep learning. However, due to the differences in the acquisition parameters and acquisition methods of different OCT acquisition devices, the distribution of data collected by different devices is different, which will seriously affect the detection results. , leading to deviations in the detection results, which in turn makes the detection accuracy low

Images