孙娜娜+刘鑫
摘 要: 图像显著性检测是一种通过对图像颜色、强度、方向等特征进行分析生成图像显著性图的技术。其生成的显著性图可以用于图像分割、图像压缩以及图像识别等图像处理领域,从而改善图像处理的性能。为了对图像显著性检测技术及其发展有一个全面深入的了解,使用文献研究法和比较研究法对其概念及方法进行了探究。针对几种具有代表性的图像显著性检测算法进行了简要的概述和分析,用流程图简明扼要地表示显著性检测算法的基本框架。研究结果显示,图像显著性检测技术的效率在不断提升,算法越来越多样化,在图像处理领域的应用越来越广泛,这些对于图像处理自动化具有重要意义。
关键词: 图像显著性; 显著性检测; 检测方法; 图像处理
中图分类号: TN919?34; TP391 文献标识码: A 文章编号: 1004?373X(2014)22?0001?04
Analysis for method of image saliency detection
SUN Na?na, LIU Xin
(School of Physical and Electronic Science, Shandong Normal University, Jinan 250014, China)
Abstract: Image saliency detection is the technology that generates image saliency diagram by means of analyzing the image features such as color, intensity and direction. The generated saliency diagram can be used in the image processing fields such as image segmentation, image compression and image recognition, so as to improve the performance of the image processing. In order to comprehend the image saliency detection technology and its development completely, its concept and methods were explored with the literature research method and comparative study method. Several image saliency detection methods with representativeness are summarized and analyzed briefly. The basic framework of saliency detection algorithm is shown in the flow chart concisely. The research results show that the efficiency of image saliency detection technology is improved constantly, its algorithm is more and more diversified, and more widely applied in the field of image processing. These have important significance for automated image processing.
Keywords: image saliency; saliency detection; detection method; image processing
0 引 言
人类所获得的外界信息80%以上都是通过视觉完成的,然而对于大量的信息,视觉系统并不是完全地进行捕获和处理,它会根据特有的机制进行选择性的处理和忽略,这就是视觉的选择注意机制。这种信息处理机制被应用于图像和视频的处理,发展出一个新的科研方向,即显著性检测研究。
随着图像处理技术和计算机技术的发展,人们越来越希望计算机能够自主地进行图像的处理,而图像显著性信息对于图像自动化处理来说异常重要。因此,图像显著性检测技术受到了广泛的关注,并产生了许多检测方法与技术。……