| 多尺度的图像显著性检测方法 |
| |
| 引用本文： | 贾宁,柳先辉,陈宇飞,赵卫东,邢尚文.多尺度的图像显著性检测方法[J].同济大学学报(自然科学版),2019,47(2):0275-0284. |
| |
| 作者姓名： | 贾宁 柳先辉 陈宇飞 赵卫东 邢尚文 |
| |
| 作者单位： | 同济大学 电子与信息工程学院， 上海 201804,同济大学 电子与信息工程学院， 上海 201804,同济大学 电子与信息工程学院， 上海 201804,同济大学 电子与信息工程学院， 上海 201804,国家电网集团, 山东 济南 250000 |
| |
| 基金项目： | 国家“八六三”高技术研究发展计划（填写项目编号）2017YFB0304102 |
| |
| 摘 要： | 为了提高显著性检测算法的准确性与鲁棒性,提出了一种基于多尺度融合的对象显著性检测方法.首先对图像进行平滑处理,过滤掉图像中的高频噪声特征,然后对图像进行尺度划分并分别采用不同的方法对不同尺度上的图像检测其显著性,最后根据条件随机场理论对不同尺度上的显著性检测结果进行加权融合,得到最终的显著性检测结果.在两种公共数据集上与多种经典算法进行定性、量化比较,结果表明该算法具有更好的表现.
|
| 关 键 词： | 显著性 多尺度融合 条件随机场 |
| 收稿时间： | 2018/4/19 0:00:00 |
| 修稿时间： | 2018/12/4 0:00:00 |
A Multiscale Image Saliency Detection Method |
| |
| JIA Ning,LIU Xianhui,CHEN Yufei,ZHAO Weidong and XING Shangwen.A Multiscale Image Saliency Detection Method[J].Journal of Tongji University(Natural Science),2019,47(2):0275-0284. |
| |
| Authors: | JIA Ning LIU Xianhui CHEN Yufei ZHAO Weidong and XING Shangwen |
| |
| Institution: | College of Electronic and Information Engineering, Tongji University, Shanghai 201804, China,College of Electronic and Information Engineering, Tongji University, Shanghai 201804, China,College of Electronic and Information Engineering, Tongji University, Shanghai 201804, China,College of Electronic and Information Engineering, Tongji University, Shanghai 201804, China and State Grid Corporation of China, Jinan 250000, China |
| |
| Abstract: | In order to improve the accuracy and robustness of the saliency detection algorithm, this paper proposed a multiscale image saliency detection method. First, the smoothing algorithm was adopted to filter out the noise characteristics in the image. Then, the multiscale representation of an image was performed and saliency maps were computed at different scales. Finally, according to the conditional random field theory, the saliency detection results at different scales were weighted together to get the final results. Extensive experiments in which the proposed method was compared with 9 existing state of the art methods on five benchmark data sets, ECSSD and MSRA10K, show that the proposed method performs better in terms of various evaluation metrics. |
| |
| Keywords: | saliency multiscale fusion conditional random field |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《同济大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《同济大学学报(自然科学版)》下载免费的PDF全文 |
|
