显著区域检测算法综述

Survey of Salient Region Detection Algorithms

检测视觉上显著的区域对于很多计算机视觉应用都是非常有帮助的,例如:内容保持的图像缩放,自适应的图像压缩和图像分割。显著区域检测成为视觉显著性检测领域的重要研究方向。文中介绍了显著区域检测算法的研究现状并分析了典型的显著区域检测方法。首先,将现有的显著区域检测算法进行了分类和分析。然后,在一个包含1 000幅图像的公开数据集上对典型的显著区域检测算法进行了评测。最后对现有的显著区域检测算法进行了总结并展望了下一步发展方向。

Detection of visually salient region is useful for many computer vision applications such as content aware image resizing, adaptive image compression and image segmentation. Salient region detection becomes an appealing research field in visual saliency detection. In this survey, the author introduces the research status of salient region detection methods and analyzes the classical approaches. First, the paper categorizes and analyzes the existing salient region detection algorithms, then evaluates the classical salient region detection algorithms on a public dataset including 1 000 images. Finally, the paper states the conclusion and discusses the future work on salient region detection.