Salient object detection dataset with adversarial attacks for genetic programming and neural networks.

Abstract

Machine learning is central to mainstream technology and outperforms classical approaches to handcrafted feature design. Aside from its learning process for artificial feature extraction, it has an end-to-end paradigm from input to output, reaching outstandingly accurate results. However, security concerns about its robustness to malicious and imperceptible perturbations have drawn attention since humans or machines can change the predictions of programs entirely. Salient object detection is a research area where deep convolutional neural networks have proven effective but whose trustworthiness represents a significant issue requiring analysis and solutions to hackers' attacks. This dataset is an image repository containing five different image databases to evaluate adversarial robustness by introducing 12 adversarial examples, each leveraging a known adversarial attack or noise perturbation. The dataset comprises 56,387 digital images, resulting from applying adversarial examples on subsets of four standard databases (i.e., FT, PASCAL-S, ImgSal, DUTS) and a fifth database (SNPL) portraying a real-world visual attention problem of a shorebird called the snowy plover. We include original and rescaled images from the five databases used with the adversarial examples as part of this dataset for easy access and distribution.