Salient target detection method based on a cascade convolutional network and adversarial learning

Abstract

The invention discloses a salient target detection method based on a cascade convolutional network and adversarial learning. The method comprises the following steps: 1, designing a global saliency estimator E; 2, designing a local saliency refiner R; 3, combining the global saliency estimator E and the local saliency refiner R into a generator G based on a cascade convolutional neural network forgenerating a saliency map; 4, optimizing the generator G; 5, designing an adversarial learning discriminator D to distinguish a real saliency map from a predicted saliency map generated by a generator G; and 6, the generator G and the adversarial learning discriminator D follow the CGAN strategy and are trained in a complete end-to-end manner, so that the generator G can better understand the structure information of the salient object, and a good saliency detection result is obtained. According to the method, the structural information is learned implicitly through confrontation learning, sothat significance target detection can be well carried out, and a best result is obtained on a plurality of databases.

Description

technical field [0001] The invention relates to a salient target detection method, in particular to a salient target detection method based on cascaded convolutional network and adversarial learning (CCAL). Background technique [0002] Saliency object detection is to locate objects that attract people's attention in natural images by assigning large saliency values ​​to some regions. With the advancement of saliency detection technology, more and more applications in the field of image processing and computer vision have begun to use the results of saliency detection to improve their performance, such as image segmentation, image cropping, object detection, image retrieval. and many more. However, ineffective saliency detection results will directly affect the performance of the above-mentioned related applications based on saliency detection, thus limiting the application scope and application effect of saliency detection methods to a certain extent. In recent years, sal...

AI Technical Summary

Problems solved by technology

However, it is still a challenge to obtain clear saliency boundaries and consistent saliency regions

Images