Training method and device of image processing model and electronic equipment

Abstract

The invention discloses an image processing model training method and device and electronic equipment, and the method comprises the steps: carrying out the multi-time iteration training of a first image processing model, and obtaining a loss ratio of the first image processing model based on an image processing model obtained through each iteration training in the multi-time iteration training and a loss ratio corresponding to each iteration training; and determining a target image processing model corresponding to the first image processing model. According to the method, after the first image processing model is obtained, the first image processing model is subjected to iteration training for multiple times, the image processing model with the most accurate prediction result is screened out from the image processing models subjected to iteration for multiple times, and the image processing model is used as the target image processing model; and the accuracy of the target image processing model is ensured.

Description

technical field [0001] The present application relates to the technical field of image recognition, in particular to a training method, device and electronic equipment for an image processing model. Background technique [0002] In the electric power overhaul operation site, in order to ensure the safety of the staff and prevent the accidents of personnel working at heights falling, the staff are required to wear seat belts in accordance with the regulations. In order to improve the efficiency of seat belt supervision, an intelligent seat belt detection system is needed. [0003] At present, in order to detect whether the staff is wearing a safety belt, the first method adopted is to distinguish the safety helmet and safety belt based on the deep convolutional neural network. The correlation model detects the staff images in the video and judges the wearing conditions of their helmets and safety belts. The image processing model is obtained based on the deep convolutional n...

AI Technical Summary

Problems solved by technology

[0003] At present, in order to detect whether the staff is wearing a safety belt, the first method adopted is to distinguish the safety helmet and seat belt detection method based on the deep convolutional neural network, through the deep convolutional neural network model and the space used to lock the safety helmet and safety belt The correlation model detects the staff images in the video and judges the wearing conditions of their helmets and safety belts. The image processing model is obtained based on the deep convolutional neural network model and the spatial correlation model. During the training process of the product neural network, the image features are extracted again by reducing the number of channels and increasing the number of channels. When the number of channels increases, more image features will be obtained, which will lead to overfitting of the image processing model during the training process. Although Overfitting will make the prediction accuracy of the image processing model training process high, but when the image processing model is used for actual prediction, the actual prediction result accuracy will decrease instead, and the image features are used to detect whether the staff wears When the seat belt is worn, the staff cannot be accurately locked, resulting in a low accuracy rate of the detection results of the image processing model

Images