基于背景评估的贝叶斯模型显著性检测

计算机与现代化 ›› 2021, Vol. 0 ›› Issue (10): 63-68.

基于背景评估的贝叶斯模型显著性检测

(1.安康学院电子与信息工程学院电子信息技术研究中心，陕西安康725000;
2.天水市绿色催化专家智库办公室，甘肃天水741400)

出版日期:2021-10-14 发布日期:2021-10-14
作者简介:文雅宏（1993—），男，甘肃天水人，讲师，硕士，研究方向：图像处理，目标检测，E-mail: 1172060962@qq.com; 巨琛（1986—），男，甘肃酒泉人，讲师，硕士，研究方向：图像处理，E-mail: 105189517@qq.com。
基金资助:
国家自然科学基金资助项目(61801005); 安康学院校级基金资助项目(2019AYQJ03); 陕西省教育厅基金资助项目(21JK0468)

A Bayesian Model Saliency Detection Algorithm Based on Background Information Evaluation

(1. Research Center of Electronic Information Technology, School of Electronic and Information Engineering,
Ankang Universily, Ankang 725000, China;

Online:2021-10-14 Published:2021-10-14

摘要/Abstract

摘要： 针对自然图像中，复杂背景信息对显著性目标检测的影响，提出一种利用背景信息进行预测和贝叶斯模型选择优化的显著性检测方法。首先，为了提取完整的先验信息，根据背景信息与图像边界的连通性，以及对图像边界是否为背景进行评估来生成先验显著图。其次，为了降低背景信息的干扰，通过对流行排序算法生成的显著图进行角点检测，选择较为准确的显著点构造凸包。最后，利用贝叶斯模型进行选择优化来抑制和显著目标具有相同特征的背景信息。在2个公开的数据集上进行测试，并与4种性能较好的显著性检测算法对比，结果显示本文算法可提高显著性检测的准确性和区域的完整性。

关键词: 显著性检测, 背景, 贝叶斯模型, 凸包

Abstract: Aiming at the influence of complex background information on salient object detection in natural images, this paper proposes a saliency detection method based on background information prediction and Bayesian model selection optimization. First, in order to extract complete prior information, a prior saliency map is generated according to the evaluation of the connectivity between the background information and the image boundary, and whether the image boundary is the background. Secondly, in order to reduce the interference of background information, corner detection is performed on saliency map generated by popular sorting algorithm, and the more accurate salieney points are selected to construct convex hull. Finally, Bayesian model is used for selection optimization to suppress the background information with the same characteristics as the salient object. Experiment is tested on two public datasets and compared with four classical saliency detection algorithms. The results show that the proposed algorithm can improve the accuracy of saliency detection and regional integrity.

Key words: saliency detection, background, Bayesian model, convex hull

文雅宏, 巨琛. 基于背景评估的贝叶斯模型显著性检测[J]. 计算机与现代化, 2021, 0(10): 63-68.

WEN Ya-hong, JU Chen. A Bayesian Model Saliency Detection Algorithm Based on Background Information Evaluation[J]. Computer and Modernization, 2021, 0(10): 63-68.

参考文献［23］

［1］	ZHANG J, LIU Y M, ZHANG S P, et al. Light field saliency detection with deep convolutional networks［J］. IEEE Transactions on Image Processing, 2020,29:4421-4434.
［2］	赵应丁,岳星宇,杨文姬,等. 基于多特征融合卷积神经网络的显著性检测［J］. 计算机工程与科学, 2021,43(4):729-737.
［3］	张艺涵,张朝晖,霍丽娜,等. 结合双流特征融合及对抗学习的图像显著性检测［J］. 计算机辅助设计与图形学学报, 2021,33(3):376-384.
［4］	LIU N, HAN J W, YANG M H. PiCANet: Pixel-wise contextual attention learning for accurate saliency detection［J］. IEEE Transactions on Image Processing, 2020,29:6438-6451.
［5］	CAI Y F, DAI L, WANG H, et al. A novel saliency detection algorithm based on adversarial learning model［J］. IEEE Transactions on Image Processing, 2020,29:4489-4504.
［6］	赵世敏,王鹏杰,曹乾,等. 基于图像语义的弱监督显著性物体检测［J］. 计算机辅助设计与图形学学报, 2021,33(2):270-277.
［7］	ITTI L, KOCH C, NIEBUR E. A model of saliency-based visual attention for rapid scene analysis［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1998,20(11):1254-1259.
［8］	LU H C, LI X H, ZHANG L H, et al. Dense and sparse reconstruction error based saliency descriptor［J］. IEEE Transactions on Image Processing, 2016,25(4):1592-1603.
［9］	CHENG M M, MITRA N J, HUANG X L, et al. Global contrast based salient region detection［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015,37(3):569-582.
［10］	CHEN D Y, JIA T, WU C D. Visual saliency detection: From space to frequency［J］. Signal Processing: Image Communication, 2016,44:57-68.
［11］	YANG C, ZHANG L H, LU H C, et al. Saliency detection via graph-based manifold ranking［C］// Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2013:3166-3173.
［12］	QIN Y, LU H C, XU Y Q, et al. Saliency detection via cellular automata［C］// Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2015:110-119.
［13］	ZHU W J, LIANG S, WEI Y C, et al. Saliency optimization from robust background detection［C］// Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2014:2814-2821.
［14］	姜青竹,田畅,吴泽民,等. 基于可区分边界和加权对比度优化的显著度检测算法［J］. 电子学报, 2017,45(1):147-156.
［15］	XIE Y L, LU H C. Visual saliency detection based on Bayesian model［C］// Proceedings of the 2011 18th IEEE International Conference on Image Processing. 2011:645-648.
［16］	RAHTU E, KANNALA J, SALO M, et al. Segmenting salient objects from images and videos［C］// Proceedings of the 11th European Conference on Computer Vision. 2010:366-379.
［17］	王万耀,段先华,徐丹,等. 基于显著性的Grabcut图像分割方法［J］. 计算机工程, 2018,44(7):230-236.
［18］	PERAZZI F, KRAHENBUHL P, PRITCH Y, et al. Saliency filters: Contrast based filtering for salient region detection［C］// Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2012:733-740.
［19］	WEI Y C, WEN F, ZHU W J, et al. Geodesic saliency using background priors［C］// Proceedings of the 12th European Conference on Computer Vision. 2012,3:29-42.
［20］	ZHOU T, FAN D P, CHENG M M, et al. RGB-D salient object detection: A survey［J］. Computational Visual Media, 2021,7(1):37-69.
［21］	BORJI A, CHENG M M, HOU Q B, et al. Salient object detection: A survey［J］. Computational Visual Media, 2019,5(2):117-150.
［22］	申扬,李巍,刚毅凝,等. 基于背景先验与低秩恢复的显著性目标检测方法［J］. 计算机与现代化, 2019(1):33-39.
［23］	魏计鹏,秦国峰. 基于深度图像增益的RGB-D显著性物体检测［J］. 计算机与现代化, 2021(5):26-30.

[1]	闫子贤, 董宝良, 唐思谜. 针对复杂背景下低分辨率舰船目标的改进YOLOv7算法[J]. 计算机与现代化, 2023, 0(11): 120-126.
[2]	刘宏飞, 杨耀权, 杨雨航. 基于改进光流特征的运动目标跟踪[J]. 计算机与现代化, 2021, 0(03): 115-121.
[3]	冯洋. 基于小波包变换的红外弱小目标检测[J]. 计算机与现代化, 2020, 0(12): 112-115.
[4]	潘陈听, 谭晓阳, . 复杂背景下基于深度学习的视频动作识别[J]. 计算机与现代化, 2020, 0(07): 97-103.
[5]	朱坤. 基于高斯混合-贝叶斯模型的轨迹预测[J]. 计算机与现代化, 2019, 0(02): 72-.
[6]	申扬1,李巍1,刚毅凝1,赵睿1,郝跃冬2,王超3. 基于背景先验与低秩恢复的显著性目标检测方法[J]. 计算机与现代化, 2019, 0(01): 33-.
[7]	崔莹,陈升东,袁峰,李引. 基于改进ViBe的室外行人检测方法[J]. 计算机与现代化, 2018, 0(03): 30-.
[8]	李亚,王颖. 基于一维最大熵的视频图像运动背景减除[J]. 计算机与现代化, 2018, 0(03): 44-.
[9]	汪洋1,2，许映秋1,2，彭艳兵1,2. 基于KNN技术的校内网验证码识别[J]. 计算机与现代化, 2017, 0(2): 93-97.
[10]	缑新科，王跃. 基于T矩阵归一化PLDA的说话人确认[J]. 计算机与现代化, 2017, 0(10): 53-56.
[11]	李慧霞1，李临生1，闫庆森2，周景文1. 基于Mean Shift算法的目标跟踪综述[J]. 计算机与现代化, 2017, 0(1): 65-70.
[12]	马洪霞1，丛林虎2. 基于改进非等间隔灰色预测模型的导弹退化状态预测[J]. 计算机与现代化, 2016, 0(3): 5-10.
[13]	焦玉清,王文中,罗斌. 基于场景语义先验和全局外观一致性的运动目标检测[J]. 计算机与现代化, 2016, 0(11): 58-63.
[14]	丁曹凯，周武能. 一种改进的基于上下文的破损显著性检测[J]. 计算机与现代化, 2015, 0(9): 42-45.
[15]	李旭，徐贵力,程月华，王彪，田裕鹏，李开宇. 一种基于像素抽样建模的车辆目标提取方法[J]. 计算机与现代化, 2014, 0(3): 89-93.