Image processing method and device and electronic device

Abstract

The embodiment of the invention provides an image processing method and device and an electronic device, and belongs to the technical field of data processing. The method comprises the steps of obtaining a first image containing a target object; setting a segmentation network for performing image processing on the first image; setting a plurality of parallel convolution layers with different sampling rates behind a second downsampling layer in the segmentation network, wherein the parallel convolution layers are used for processing images outputted by the second downsampling layer, and the image features extracted from each parallel convolution layer form a second image in a fusion manner; and obtaining a third image containing the target object by carrying out the target recognition on the second image. Through the scheme of the invention, the accuracy of target identification is improved. Through the processing scheme disclosed by the invention, the stylization effect of the image can be set in real time.

Description

technical field [0001] The present disclosure relates to the technical field of data processing, and in particular to an image processing method, device and electronic equipment. Background technique [0002] With the development of artificial intelligence technology, more and more image processing tasks can be done through artificial intelligence. As a means of artificial intelligence, neural network has been fully applied in the field of computer image recognition. For example, the recognition of different people in images, or the automatic recognition of different objects on the road in driverless driving. These constitute the specific content of image semantic recognition. The process of image semantic recognition involves image semantic segmentation. Image semantic segmentation is generally modeled as a pixel-level multi-classification problem, and its goal is to distinguish each pixel of the image into one of several predefined categories. [0003] Most of the existi...

AI Technical Summary

Problems solved by technology

However, although this network structure can obtain better semantic segmentation results, once the codec structure is adopted, the spatial resolution of the feature map will be significantly reduced during the encoding process. Although the original resolution of the image is restored during the upsampling process, but It will inevitably lead to the loss of spatial detail information, which will lead to a decrease in the accuracy of target recognition

Images