Image recognition method and device and computer readable medium

Abstract

The invention provides an image recognition method and device and a computer readable medium. According to the image recognition method in the embodiment, an artificial neural network is utilized to recognize whether a target image is contained in an original image. The method comprises the steps of contour information acquisition operation (S1), wherein a dimension reduction algorithm of the artificial neural network is utilized to acquire global contour information of the original image from the original image; target image region acquisition operation (S2), wherein according to the global contour information and a sample of the target image, an image region similar to the target image is acquired from the original image to serve as a target image region; and image recognition operation(S3), wherein according to the acquired target image region, the sample of the target image is utilized to perform image recognition based on a dimension rise algorithm of the artificial neural network.

Description

Technical field [0001] The present disclosure relates to the field of computer technology, and more specifically, to an image recognition method and device using artificial neural network technology, and a computer-readable medium. Background technique [0002] With the rapid development of computer technology, especially artificial intelligence, artificial neural network technology is increasingly used in many technical fields. In recent years, especially in image recognition technology, artificial neural network technology (for example, bp neural network technology, deep learning technology, template matching technology, etc.) is continuously being widely used. [0003] However, in the process of realizing the present invention, the inventors found that at least the following problems exist in the prior art. For example, in the existing artificial neural network technology applied to image recognition, because prior knowledge is not considered, it is easy to use image details as...

AI Technical Summary

Problems solved by technology

For example, in the existing artificial neural network technology applied to image recognition, because prior knowledge is not considered, it is easy to use image details as features, which sometimes leads to image overfitting and affects the accuracy of image recognition

Moreover, image recognition based on artificial neural networks (especially fully connected neural networks) for the entire original image will generate a large amount of computation.

Images