基于加權(quán)的K近鄰線性混合顯著性目標(biāo)檢測
doi: 10.11999/JEIT190093
-
安徽大學(xué)計算機(jī)科學(xué)與技術(shù)學(xué)院 合肥 230601
Salient Object Detection Using Weighted K-nearest Neighbor Linear Blending
-
College of Computer Science and Technology, Anhui University, Hefei 230601, China
-
摘要: 顯著性目標(biāo)檢測旨在于一個場景中自動檢測能夠引起人類注意的目標(biāo)或區(qū)域，在自底向上的方法中，基于多核支持向量機(jī)(SVM)的集成學(xué)習(xí)取得了卓越的效果。然而，針對每一張要處理的圖像，該方法都要重新訓(xùn)練，每一次訓(xùn)練都非常耗時。因此，該文提出一個基于加權(quán)的K近鄰線性混合(WKNNLB)顯著性目標(biāo)檢測方法：利用現(xiàn)有的方法來產(chǎn)生初始的弱顯著圖并獲得訓(xùn)練樣本，引入加權(quán)的K近鄰(WKNN)模型來預(yù)測樣本的顯著性值，該模型不需要任何訓(xùn)練過程，僅需選擇一個最優(yōu)的K值和計算與測試樣本最近的K個訓(xùn)練樣本的歐式距離。為了減少選擇K值帶來的影響，多個加權(quán)的K近鄰模型通過線性混合的方式融合來產(chǎn)生強的顯著圖。最后，將多尺度的弱顯著圖和強顯著圖融合來進(jìn)一步提高檢測效果。在常用的ASD和復(fù)雜的DUT-OMRON數(shù)據(jù)集上的實驗結(jié)果表明了該算法在運行時間和性能上的有效性和優(yōu)越性。當(dāng)采用較好的弱顯著圖時，該算法能夠取得更好的效果。
-
關(guān)鍵詞:
- 顯著性目標(biāo)檢測 /
- 集成學(xué)習(xí) /
- 線性混合 /
- 加權(quán)的K近鄰
Abstract: Salient object detection which aims at automatically detecting what attracts human’s attention most in a scene, bootstrap learning based on Support Vector Machine(SVM) has achieved excellent performance in bottom-up methods. However, it is time-consuming for each image to be trained once based on multiple kernel SVM ensemble. So a salient object detection model via Weighted K-Nearest Neighbor Linear Blending (WKNNLB) is proposed. First of all, existing saliency detection methods are employed to generate weak saliency maps and obtain training samples. Then, Weighted K-Nearest Neighbor (WKNN) is introduced to learning salient score of samples. WKNN model needs no pre-training process, only needs selecting K value and computing saliency value by the K-nearest neighbors labels of training sample and the distances between the K-nearest neighbors training samples and the testing sample. In order to reduce the influence of selecting K value, linear blending of multi-WKNNs is applied to generating strong saliency maps. Finally, multi-scale saliency maps of weak and strong model are integrated together to further improve the detection performance. The experimental results on common ASD and complex DUT-OMRON datasets show that the algorithm is effective and superior in running time and performance. It can even perform favorable against the state-of-the-art methods when adopting better weak saliency map. -
表 1 特征
fji 取值(65維)特征維度序號 特征 維度大小 取值范圍 0～2 平均RGB值 3 [0,1] 3～5 平均CIELab值 3 [0,1] 6～64 LBP直方圖值 59 [0,1] 
下載: 導(dǎo)出CSV
表 2 5種經(jīng)典方法及其提高在F-度量值的對比
算法 IT IT+ LRMR LRMR+ GC GC+ MR MR+ MBD MBD+ ASD 0.381 0.542 0.727 0.829 0.811 0.848 0.869 0.876 0.855 0.867 DUT-OMRON 0.343 0.541 0.498 0.545 0.520 0.554 0.574 0.576 0.850 0.854
下載: 導(dǎo)出CSV
表 3 WKNNLB和BLSVM在4個數(shù)據(jù)集上F-度量和運行時間(s)對比
ASD SED2 PASCAL-S DUT-OMRON F-measure Time F-measure Time F-measure Time F-measure Time WKNNLB 0.820 4058 0.758 332 0.655 5000 0.534 30864 BLSVM 0.810 8093 0.740 720 0.651 11184 0.524 65120
下載: 導(dǎo)出CSV
-
BORJI A and ITTI L. State-of-the-art in visual attention modeling[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(1): 185–207. doi: 10.1109/TPAMI.2012.89 ITTI L. Automatic foveation for video compression using a neurobiological model of visual attention[J]. IEEE Transactions on Image Processing, 2004, 13(10): 1304–1318. doi: 10.1109/TIP.2004.834657 ZHANG Fan, DU Bo, and ZHANG Liangpei. Saliency-guided unsupervised feature learning for scene classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(4): 2175–2184. doi: 10.1109/TGRS.2014.2357078 LU Xiaoqiang, ZHENG Xiangtao, and LI Xuelong. Latent semantic minimal hashing for image retrieval[J]. IEEE Transactions on Image Processing, 2017, 26(1): 355–368. doi: 10.1109/TIP.2016.2627801 WEI Yunchao, XIAO Huaxin, SHI Honghui, et al. Revisiting dilated convolution: A simple approach for weakly-and semi-supervised semantic segmentation[C]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 7268–7277. doi: 10.1109/CVPR.2018.00759. ZHANG Xiaoning, WANG Tiantian, QI Jinqing, et al. Progressive attention guided recurrent network for salient object detection[C]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 714–722. doi: 10.1109/CVPR.2018.00081. CHEN Shuhan, TAN Xiuli, WANG Ben, et al. Reverse attention for salient object detection[C]. The 15th European Conference on Computer Vision, Munich, Germany, 2018: 236–252. doi: 10.1007/978-3-030-01240-3_15. ZHANG Lu, DAI Ju, LU Huchuan, et al. A bi-directional message passing model for salient object detection[C]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 1741–750. doi: 10.1109/CVPR.2018.00187. WANG Tiantian, ZHANG Lihe, WANG Shuo, et al. Detect globally, refine locally: A novel approach to saliency detection[C]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 3127–3135. doi: 10.1109/CVPR.2018.00330. HOU Qibin, CHENG Mingming, HU Xiaowei, et al. Deeply supervised salient object detection with short connections[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 41(4): 815–828. doi: 10.1109/TPAMI.2018.2815688 YANG Chuan, ZHANG Lihe, LU Huchuan, et al. Saliency detection via graph-based manifold ranking[C]. 2013 IEEE Conference on Computer Vision and Pattern Recognition, Portland, USA, 2013: 3166–3173. doi: 10.1109/CVPR.2013.407. CHENG Mingming, WARRELL J, LIN Wenyan, et al. Efficient salient region detection with soft image abstraction[C]. 2013 IEEE International Conference on Computer Vision, Sydney, Australia, 2013: 1529–1536. doi: 10.1109/ICCV.2013.193. ZHANG Jianming, SCLAROFF S, LIN Zhe, et al. Minimum barrier salient object detection at 80 FPS[C]. 2015 IEEE International Conference on Computer Vision, Santiago, Chile, 2015: 1404–1412. doi: 10.1109/ICCV.2015.165. BORJI A, CHENG Mingming, JIANG Huaizu, et al. Salient object detection: A benchmark[J]. IEEE Transactions on Image Processing, 2015, 24(12): 5706–5722. doi: 10.1109/TIP.2015.2487833 TONG Na, LU Huchuan, RUAN Xiang, et al. Salient object detection via bootstrap learning[C]. 2015 IEEE Conference on Computer Vision and Pattern Recognition, Boston, USA, 2015: 1884–1892. doi: 10.1109/CVPR.2015.7298798. LU Huchuan, ZHANG Xiaoning, Qi Jinqing, et al. Co-bootstrapping saliency[J]. IEEE Transactions on Image Processing, 2017, 26(1): 414–425. doi: 10.1109/TIP.2016.2627804 SONG Hangke, LIU Zhi, DU Huan, et al. Depth-aware salient object detection and segmentation via multiscale discriminative saliency fusion and bootstrap learning[J]. IEEE Transactions on Image Processing, 2017, 26(9): 4204–4216. doi: 10.1109/TIP.2017.2711277 WU Lishan, LIU Zhi, SONG Hangke, et al. RGBD co-saliency detection via multiple kernel boosting and fusion[J]. Multimedia Tools and Applications, 2018, 77(16): 21185–21199. doi: 10.1007/s11042-017-5576-y ACHANTA R, SHAJI A, SMITH K, et al. SLIC superpixels compared to state-of-the-art superpixel methods[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(11): 2274–2282. doi: 10.1109/TPAMI.2012.120 OJALA T, PIETIKAINEN M, and MAENPAA T. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(7): 971–987. doi: 10.1109/tpami.2002.1017623 ACHANTA R, HEMAMI S, ESTRADA F, et al. Frequency-tuned salient region detection[C]. 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, USA, 2009: 1597–1604. doi: 10.1109/CVPR.2009.5206596. ITTI L, KOCH C, and 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. doi: 10.1109/34.730558 SHEN Xiaohui and WU Ying. A unified approach to salient object detection via low rank matrix recovery[C]. 2012 IEEE Conference on Computer Vision and Pattern Recognition, Providence, USA, 2012: 853–860. doi: 10.1109/CVPR.2012.6247758. -
圖(8) / 表(3)
計量
- 文章訪問數(shù): 2228
- HTML全文瀏覽量: 1230
- PDF下載量: 105
- 被引次數(shù): 0
下載:
