Visual saliency detection method combined with image classification

Abstract

The invention provides a visual saliency detection method combined with image classification. The method comprises the steps of utilizing a visual saliency detecting model which comprises an image coding network, an image decoding network and an image identification model, using a multidirectional image as an input of the image coding network, and extracting an image characteristic on the condition of multiple resolution as a coding characteristic vector F; fixing a weight except for the last two layers in the image coding network, and training network parameters for obtaining a visual saliency picture of an original image; using the F as the input of the image decoding network, and performing normalization processing on the saliency picture which corresponds with the original image; for the input F of the image decoding network, finally obtaining a generated visual saliency picture through an upsampling layer and a nonlinear sigmoid layer; by means of the image identification network, using the visual saliency picture of the original image and the generated visual saliency picture as the input, performing characteristic extraction by means of a convolutional layer with a small convolution kernel and performing pooling processing, and finally outputting probability distribution of the generated picture and probability distribution of classification labels by means of three total connecting layers. The method provided by the invention realizes quick and effective image analysis and determining and furthermore realizes good effects such as saving manpower and physical resource costs and remarkably improving accuracy in practices such as image marking, supervising and behavior predicating.

Description

technical field [0001] The invention belongs to the technical field of image detection and intelligent recognition, in particular to a visual salience detection method. It is used to solve technical problems such as image annotation, supervision and behavior prediction. Background technique [0002] In an image, visual saliency detection aims to estimate the location of the most interesting object in the image. The application of visual saliency detection is very extensive in daily social life. For example, in complex shopping mall street scenes, using monitoring equipment to monitor the safety and order of shopping mall streets requires supervision and investigation of people carrying high-risk items and suspicious behaviors. In addition to on-site inspections by professionally qualified personnel, this also requires constant video monitoring. This kind of supervision and investigation not only consumes a lot of resources, but also cannot avoid major omissions caused by p...

AI Technical Summary

Problems solved by technology

[0012] The algorithm of deep learning, because the deep neural network has a strong ability of autonomous feature learning and extraction, and has the processing of the dimension of the seen features, it removes redundant information to a large extent, and through effective supervised or semi-supervised learning, It has greatly improved the practicability and stability of the algorithm, but the current image saliency detection algorithm based on deep learning is limited by the resolution, size and error transfer function of the saliency target, so there are still many problems in the algorithm

In particular, there are many false positives in target detection due to the lack of classification

[0013] It can be seen that the current image saliency detection methods have certain defects.

Images