Image main target detection method and device based on convolutional neural network

Abstract

The embodiment of the invention relates to the field of artificial intelligence, and provides an image main target detection method and device based on a convolutional neural network. The method comprises the steps of obtaining an input image; carrying out convolution layer processing on the input image to obtain a plurality of feature maps, and outputting the feature maps through a feature channel; according to the thermodynamic diagram of each feature channel, obtaining a saliency region on the feature diagram; carrying out clustering processing on the salient regions on the feature map to form a plurality of feature clusters; and determining the position of a main target on the input image based on the positions of all the feature clusters. According to the technical scheme, the problems that in the prior art, during model training, a specific target of an image needs to be manually labeled, and a trained model is only suitable for the specific target are solved.

Description

[0001] 【Technical field】 [0002] The present invention relates to the field of artificial intelligence, in particular to a convolutional neural network-based image main target detection method and device. [0003] 【Background technique】 [0004] With the development of deep learning technology, it has been widely used in visual recognition, image recognition and other fields. [0005] In the prior art, the specific target detection of an image adopts the following steps: [0006] Step 1. Obtain a large number of images containing a specific target, and mark the position information of the specific target on the image by manpower. For example, a bounding box is used to represent, and the four-dimensional vector P(x, y, w, h) of the bounding box is given to represent the position and size of the box, and then the model is obtained through training. [0007] Step 2. Use the trained model to extract specific objects in the newly acquired image. [0008] However, the existing te...

AI Technical Summary

Problems solved by technology

[0010] In view of this, the embodiment of the present invention provides a method and device for image main target detection based on convolutional neural network, which is used to solve the problem that the specific target of the image needs to be manually marked during model training in the prior art, and the trained model only needs to Questions for Specific Purposes

Images