Diffusion type atmosphere unorganized monitor mask detection method based on model training

The invention provides a diffusion type atmosphere unorganized monitor mask detection method based on model training. The method comprises the following steps: S1, generating a training sample; s2, extracting features including data complexity features and temperature and humidity correlation; s3, the mask detection model is trained, field data of X days are collected, data of the first X-10 days are used as a training set, data of the last 10 days are used as a verification set, and the classification precision of the training set reaches 95%-98% through evaluation indexes; and S4, deploying the model trained in the step S3 in a platform used by the diffusion type atmosphere unorganized monitor to carry out mask detection early warning. According to the method, on-site deployment is carried out on the diffusion type atmosphere unorganized monitor simulation masks with enough sample quantity, on-site data including monitoring factors, temperature and humidity are collected, feature extraction, model training