区域特征检测还可利用计算机技术中的树理论 进行稳定特征提取,Xu 等人[68]提出一种基于该理 论的拓扑方法 TBMR( tree-based Morse regions) 。该 方法以 Morse 理论为基础定义临界点: 最大值点、最小值点和鞍点,分别对应最大树叶子节点、最小树叶 子节点和分叉节点。TBMR 区域对应树中具有唯一 子节点和至少具有一个兄弟节点的节点。如图 10 所示,节点 A 和 C 代表最小值区域; 节点 H 和 E 代 表最大值区域; 节点 A ∪ B ∪ C ∪ D ∪ G 和 E ∪ F ∪ G ∪ H 表示鞍点区域; 节点 A ∪ B 、C ∪ D 、E ∪ F 为所求 TBMR 区域。该方法仅依赖拓扑信息,完全继承形状空间不变性,对视角变化具有鲁棒性,计算 速度快,与 MSER 具有相同复杂度,常用于图像配准 和 3 维重建。
[1] Harris C,Stephens M. A combined corner and edge detector [C]/ /Proceedings of the 4th Alvey Vision Conference. Manchester: AVC,1988: 147-151. [DOI: 10. 5244 /C. 2. 23]
[2] Rosten E,Drummond T. Machine learning for high-speed corner detection[C]/ /Proceedings of the 9th European Conference on Computer Vision. Graz,Austria: Springer,2006: 430-443. [DOI: 10. 1007 /11744023_34]
[3] Lowe D G. Distinctive image features from scale-invariantkeypoints[J]. International Journal of Computer Vision,2004, 60( 2) : 91-110. [DOI: 10. 1023 /B: VISI. 0000029664. 99615. 94]
[4] Liu L,Zhan Y Y,Luo Y,et al. Summarization of the scale invariant feature transform[J]. Journal of Image and Graphics, 2013,18( 8) : 885-892. [刘立,詹茵茵,罗扬,等. 尺度不 变特征 变 换 算 子 综 述[J]. 中 国 图 象 图 形 学 报,2013, 18( 8) : 885-892.][DOI: 10. 11834 /jig. 20130801]
[5] Xu Y X,Chen F. Recent advances in local image descriptor[J]. Journal of Image and Graphics,2015,20( 9) : 1133-1150. [许 允喜,陈方. 局部图像描述符最新研究进展[J]. 中国图象 图形学报,2015,20( 9) : 1133-1150.][DOI: 10. 11834 /jig. 20150901]
[6] Zhang X H,Li B,Yang D. A novel Harris multi-scale corner detection algorithm[J]. Journal of Electronics and Information Technology,2007,29 ( 7) : 1735-1738. [张小 洪,李 博,杨 丹. 一种新的 Harris 多尺度角点检测[J]. 电子与信息学报, 2007,29 ( 7 ) : 1735-1738.] [DOI: 10. 3724 / SP. J. 1146. 2005. 01332]
[7] He H Q,Huang S X. Improved algorithm for Harris rapid subpixel corners detection[J]. Journal of Image and Graphics, 2012,17( 7) : 853-857. [何海清,黄声享. 改进的 Harris 亚 像素角点快速定位[J]. 中国图象图形学报,2012,17( 7) : 853-857.][DOI: 10. 11834 /jig. 20120715]
[8] Zhang L T,Huang X L,Lu L L,et al. Fast Harris corner detection based on gray difference and template[J]. Chinese Journal of Scientific Instrument,2018,39( 2) : 218-224. [张立亭,黄 晓浪,鹿琳琳,等. 基于灰度差分与模板的 Harris 角点检测 快速算法[J]. 仪器仪表学报,2018,39( 2) : 218-224.]
[9] Ke Y,Sukthankar R. PCA-SIFT: a more distinctive representation for local image descriptors[C]/ /Proceedings of 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington,DC: IEEE,2004: 506-513. [DOI: 10. 1109 /CVPR. 2004. 1315206]
[10] Bay H,Tuytelaars T,Gool L. SURF: speeded up robust features [C]/ /Proceedings of the 9th European Conference on Computer Vision. Graz,Austria: Springer,2006: 404-417. [DOI: 10. 1007 /11744023_32]
[11] Liu L,Peng F Y,Zhao K,et al. Simplified SIFT algorithm for fast image matching[J]. Infrared and Laser Engineering,2008, 37( 1) : 181-184. [刘立,彭复员,赵坤,等. 采用简化 SIFT 算法实 现 快 速 图 像 匹 配[J]. 红外与激光工程,2008, 37( 1) : 181-184.][DOI: 10. 3969 /j. issn. 1007-2276. 2008. 01. 042]
[12] Abdel-Hakim A E,Farag A A. CSIFT: a SIFT descriptor with color invariant characteristics[C]/ /Proceedings of 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York,NY: IEEE,2006: 1978-1983. [DOI: 10. 1109 /CVPR. 2006. 95]
[13] Mikolajczyk K,Schmid C. A performance evaluation of local descriptors[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27 ( 10 ) : 1615-1630. [DOI: 10. 1109 /TPAMI. 2005. 188]
[14] Morel J M,Yu G S. ASIFT: a new framework for fully affine invariant image comparison[J]. SIAM Journal on Imaging Sciences,2009,2( 2) : 438-469. [DOI: 10. 1137 /080732730]
[15] Rosten E,Porter R,Drummond T. Faster and better: a machine learning approach to corner detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32( 1) : 105- 119. [DOI: 10. 1109 /TPAMI. 2008. 275]
[16] Verdie Y,Yi K M,Fua P,et al. TILDE: a temporally invariant learned DEtector[C]/ /Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston,MA: IEEE, 2015: 5279-5288. [DOI: 10. 1109 /CVPR. 2015. 7299165]
[17] Zhang X,Yu F X,Karaman S,et al. Learning discriminative and transformation covariant local feature detectors[C]/ /Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu,HI: IEEE,2017: 4923-4931. [DOI: 10. 1109 /CVPR. 2017. 523]
[18] Savinov N,Seki A,Ladicky L,et al. Quad-networks: unsupervised learning to rank for interest point detection[C]/ /Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu,HI: IEEE,2017: 3929-3937. [DOI: 10. 1109 /CVPR. 2017. 418]
[19] Simo-Serra E,Trulls E,Ferraz L,et al. Discriminative learning of deep convolutional feature point descriptors[C]/ /Proceedings of 2015 IEEE International Conference on Computer Vision. Santiago,Chile: IEEE,2015: 118-126. [DOI: 10. 1109 / ICCV. 2015. 22]
[20] Yi K M,Trulls E,Lepetit V,et al. LIFT: learned invariant feature transform[C]/ /Proceedings of the 14th European Conference on Computer Vision. Amsterdam,The Netherlands: Springer,2016: 467-483. [DOI: 10. 1007 /978-3-319-46466-4_28]
[21] Jaderberg M,Simonyan K,Zisserman A,et al. Spatial transformer networks[C]/ /Proceedings of the 28th International Conference on Neural Information Processing Systems. Montreal, Canada: ACM,2015: 2017-2025.
[22] Yi K M,Verdie Y,Fua P,et al. Learning to assign orientations to feature points[C]/ /Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas,NV: IEEE,2016: 107-116. [DOI: 10. 1109 /CVPR. 2016. 19]
[23] Liu C,Yuen J,Torralba A. SIFT flow: dense correspondence across scenes and its applications[J]. IEEE Transactions on Pattern Analysisand Machine Intelligence,2011,33( 5) : 978-994. [DOI: 10. 1109 /TPAMI. 2010. 147]
[24] Bristow H,Valmadre J,Lucey S. Dense semantic correspondence where every pixel is a classifier[C]/ /Proceedings of 2015 IEEE International Conference on Computer Vision. Santiago, Chile: IEEE,2015: 4024-4031. [DOI: 10. 1109 / ICCV. 2015. 458]
[25] Novotny D,Larlus D,Vedaldi A. AnchorNet: A weakly supervised network to learn geometry-sensitive features for semantic matching[C]/ /Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu,HI: IEEE, 2017: 2867-2876. [DOI: 10. 1109 /CVPR. 2017. 306]
[26] Kar A,Tulsiani S,Carreira J,et al. Category-specific object reconstruction from a single image[C]/ /Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, MA: IEEE,2015: 1966-1974. [DOI: 10. 1109 /CVPR. 2015. 7298807]
[27] Thewlis J,Bilen H,Vedaldi A. Unsupervised learning of object landmarks by factorized spatial embeddings[C]/ /Proceedings of 2017 IEEE International Conference on Computer Vision. Venice,Italy: IEEE,2017: 3229-3238. [DOI: 10. 1109 / ICCV. 2017. 348]
[28] Wang Q Q,Zhou X W,Daniilidis K. Multi-image semantic matching by mining consistent features[C]/ /Proceedings of 2018 IEEE /CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City,UT: IEEE,2018: 685-694. [DOI: 10. 1109 /CVPR. 2018. 00078]
[29] Yu D D,Yang F,Yang C Y,et al. Fast rotation-free featurebased image registration using improved N-SIFT and GMM-based parallel optimization[J]. IEEE Transactions on Biomedical Engineering,2016,63 ( 8) : 1653-1664. [DOI: 10. 1109 /TBME. 2015. 2465855]
[30] Pock T,Urschler M,Zach C,et al. A duality based algorithm for TV - L1 - optical-flow image registration[C]/ /Proceedings of the 10th International Conference on Medical Image Computing and Computer-Assisted Intervention. Brisbane,Australia: Springer, 2007: 511-518. [DOI: 10. 1007 /978-3-540-75759-7_62]
[31] Zhang G M,Sun X X,Liu J X,et al. Research on TV-L1 optical flow model for image registration based on fractional-order differentiation[J]. Acta Automatica Sinica,2017,43 ( 12) : 2213- 2224. [张桂梅,孙晓旭,刘建新,等. 基于分数阶微分的 TV-L1光流 模 型 的 图 像 配 准 方 法 研 究[J]. 自 动 化 学 报, 2017,43 ( 12 ) : 2213-2224.][DOI: 0. 16383 /j. aas. 2017. c160367]
[32] Lu X S,Tu S X,Zhang S. A metric method using multidimensional features for nonrigid registration of medical images[J]. Acta Automatica Sinica,2016,42( 9) : 1413-1420. [陆雪松, 涂圣贤,张素. 一种面向医学图像非刚性配准的多维特征度 量方法[J]. 自动化学报,2016,42( 9) : 1413-1420.][DOI: 10. 16383 /j. aas. 2016. c150608]
[33] Yang W,Zhong L M,Chen Y,et al. Predicting CT image from MRI data through feature matching with learned nonlinear local descriptors[J]. IEEE Transactions on Medical Imaging,2018, 37( 4) : 977-987. [DOI: 10. 1109 /TMI. 2018. 2790962]
[34] Cao X H,Yang J H,Gao Y Z,et al. Region-adaptive deformable registration of CT /MRI pelvic images via learning-based image synthesis[J]. IEEE Transactions on Image Processing, 2018,27 ( 7 ) : 3500-3512. [DOI: 10. 1109 /TIP. 2018. 2820424]
[35] He M M,Guo Q,Li A,et al. Automatic fast feature-level image registration for high-resolution remote sensing images[J]. Journal of Remote Sensing,2018,22( 2) : 277-292. [何梦梦,郭擎, 李安,等. 特征级高分辨率遥感图像快速自动配准[J]. 遥 感 学 报,2018,22 ( 2 ) : 277-292.] [DOI: 10. 11834 /jrs. 20186420]
[36] Fischler M A,Bolles R C. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography[J]. Communications of the ACM,1981, 24( 6) : 381-395. [DOI: 10. 1145 /358669. 358692]
[37] Torr P H S,Zisserman A. MLESAC: a new robust estimat