Sea-surface target saliency detection method based on convolutional neural network

Abstract

The invention discloses a sea-surface target saliency detection method based on a convolutional neural network. The sea-surface target saliency detection method comprises the steps of: s1, inputting an image I to be subjected to saliency detection into a trained convolutional neural network structure to obtain a multi-scale feature map of the image, and performing saliency calculation on the multi-scale feature map to obtain a corresponding first salient map; S2, meanwhile, performing color space conversion on the image I to be subjected to saliency detection to obtain a channel image, and performing saliency calculation in a spatial domain on the channel image to obtain a corresponding second salient map; S3, and fusing the first salient map and the second salient map to obtain a final salient map. The sea-surface target saliency detection method can improve the problems of insufficient feature extraction and poor robust detection result existing in the present saliency detection, sothat the saliency of a saliency region is significantly improved, and the precision of the salient result is improved.

Description

technical field [0001] The invention relates to a target detection method, in particular to a sea surface target saliency detection method based on a convolutional neural network. Background technique [0002] The detection and identification of sea surface targets are of great significance in both civil and military fields. Nowadays, there are problems of excessive information and redundant information in the rich sea surface image resources. How to remove redundancy from massive images and find needed and interesting information, saliency detection is a key to solve this problem. Naturally, the technology has attracted widespread attention. Its core is to extract the areas that attract visual attention in the image, which can provide machine-aided vision for maritime surveillance, maritime search and rescue, maritime law enforcement, ship detection, etc., and provide a basis for improving subsequent image processing and decision-making. [0003] Affected by uncertain sea ...

AI Technical Summary

Problems solved by technology

[0004] The object of the present invention is to provide a sea surface object saliency detection method based on convolutional neural network, which introduces the convolutional neural network into the sea surface object saliency detection process, and simulates the accumulation of surrounding things by the human visual system through prior training , and then simulate the human visual system to judge the region of interest in the image, automatically learn appropriate features for saliency detection, and improve the existing saliency detection problems of insufficient feature extraction and poor robust detection effect , so that the salience of the saliency area is improved, and the accuracy of the saliency result is improved

Images