Image detection method and device, computer readable storage medium and computer equipment

Abstract

The embodiment of the invention discloses an image detection method and device, a computer readable storage medium and computer equipment. The method comprises the steps of obtaining training sample data; respectively inputting the sample image into at least two neural network models to obtain an output fuzzy probability value set; calculating to obtain a loss parameter of each sample image according to the fuzzy probability value set and the label information; selecting a target sample image from the plurality of sample images according to the distribution of the loss parameters, and updating the at least two neural network models based on the target sample image; returning to execute the above steps until the at least two neural network models converge to obtain at least two trained neural network models; and detecting the to-be-detected image by using the trained at least two neural network models to obtain a detection result. Therefore, according to the method, the machine learning technology is adopted, the noise samples are screened through multi-model cooperative training, the model training effect is improved, and then the accuracy of image detection is improved.

Description

Technical field [0001] The present invention relates to the field of image processing, and more particularly to an image detecting method, a device, a computer readable storage medium, and a computer device. Background technique [0002] Convolutional Neural Networks (CNN) is a type of feedforward neural network, which contains convolutional calculations and has a depth structure, is one of the representative algorithms of depth learning (DL). The convolutional neural network has the ability to characterize the Learning, RL. It is possible to translate the input information according to its class structure, which is also referred to as "Shift-Invariant Artificial Neural NetWorks, Siann). ). [0003] In recent years, the development of convolutional neural network related technologies has developed rapidly and applications. For example, in a scenario detected by the image, the image detection model can be built using a convolutional neural network, and the efficiency of the image ...

AI Technical Summary

Problems solved by technology

[0004] However, in the current image artifact detection model built using convolutional neural networks, the artifact labels of the training sample images used in the model training stage are simple binary labels, and the inaccurate binary labels of the artifact images will affect the trained model. performance, which leads to low accuracy of image blurring detection

Images