Image processing method, device and equipment and storage medium

Abstract

The embodiment of the invention discloses an image processing method, device and equipment and a storage medium, and belongs to the technical field of image processing. The method is applied to the electronic equipment. The electronic equipment can be used for operating a neural network model, wherein the neural network model comprises a convolution layer and a normalization layer, and the normalization layer is located behind the convolution layer. The method comprises the following steps: in the process of training the neural network model, determining the number of channels of a plurality of first feature maps output by an upper layer of the normalization layer, wherein the plurality of first feature maps are at least subjected to convolution processing by the convolution layer; determining a plurality of groups of first values based on the number of channels and the plurality of first feature maps, wherein each group of first values comprises pixel values of the plurality of firstfeature maps on at least two channels, and the channels included in each group are different from each other; and performing normalization processing on the plurality of first feature maps based on the plurality of groups of first numerical values. According to the embodiment of the invention, the method can guarantee the data statistics of normalization processing, enables the precision to be higher, and guarantees the image processing effect.

Description

technical field [0001] The embodiments of the present application relate to the technical field of image processing, and in particular, to an image processing method, device, device, and storage medium. Background technique [0002] At present, the neural network model is widely used in the field of image processing technology, for example, it can be used for image recognition or image detection. A neural network model usually includes a convolutional layer, a pooling layer, a normalization layer, etc., wherein the normalization layer can be used to normalize the feature map. [0003] In related technologies, the normalization layer mostly adopts batch normalization processing. However, the batch normalization process is sensitive to the size of the batch. When the batch is small, such as the number of input feature maps is small, the statistics of the mean and variance are not accurate enough, making the accuracy low, resulting in poor image processing effect. . Content...

AI Technical Summary

Problems solved by technology

[0004] The embodiment of the present application provides an image processing method, device, equipment and storage medium, which can solve the problem of poor image processing effect due to inaccurate statistics of related technologies

Images