Abstract
This article proposes a bottom-up visual saliency model that uses the wavelet transform to conduct multiscale analysis and computation in the frequency domain. First, we compute the multiscale magnitude spectra by performing a wavelet transform to decompose the magnitude spectrum of the discrete cosine coefficients of an input image. Next, we obtain multiple saliency maps of different spatial scales through an inverse transformation from the frequency domain to the spatial domain, which utilizes the discrete cosine magnitude spectra after multiscale wavelet decomposition. Then, we employ an evaluation function to automatically select the two best multiscale saliency maps. A final saliency map is generated via an adaptive integration of the two selected multiscale saliency maps. The proposed model is fast, efficient, and can simultaneously detect salient regions or objects of different sizes. It outperforms state-of-the-art bottom-up saliency approaches in the experiments of psychophysical consistency, eye fixation prediction, and saliency detection for natural images. In addition, the proposed model is applied to automatic ship detection in optical satellite images. Ship detection tests on satellite data of visual optical spectrum not only demonstrate our saliency model's effectiveness in detecting small and large salient targets but also verify its robustness against various sea background disturbances.
Highlights
In the human neural system, a mechanism called selective visual attention has been evolved to facilitate our visual perception to rapidly locate the most important regions in a cluttered scene
To make the frequency domain model have better detection ability for both large and small salient targets, in this article, we propose a bottom-up visual saliency model based on multiscale analysis and computation in the frequency domain
To make the frequency domain model have better detection ability for both large and small salient targets, in this work, we propose a bottom-up visual saliency model based on multiscale analysis and computation in the frequency domain
Summary
In the human neural system, a mechanism called selective visual attention has been evolved to facilitate our visual perception to rapidly locate the most important regions in a cluttered scene. The proposed model performs multiscale wavelet analysis and computation in the cosine transform domain It can generate multiscale saliency maps of the scene under view. To make the frequency domain model have better detection ability for both large and small salient targets, in this work, we propose a bottom-up visual saliency model based on multiscale analysis and computation in the frequency domain. Through such a multiscale decomposition and reconstruction operation upon the magnitude coefficients in the DCT domain, our model simulates the cortical center-surround or iso-feature suppression of various scales in the spatial domain For this reason, our model can compute the multiscale saliency information simultaneously, which is very helpful to detect salient objects of different sizes. The Recall and the Precision metrics for a binary map can be calculated as Recall
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call