北京生物醫(yī)學(xué)工程

光學(xué)定位技術(shù)輔助的手術(shù)場景全景拼接算法

Panoramic stitching of surgical scenes assisted by optical positioning algorithm

作者：楊巧玲葉燦梁楠武博張楠

單位：首都醫(yī)科大學(xué)生物醫(yī)學(xué)工程學(xué)院（北京 100069） 首都醫(yī)科大學(xué)臨床生物力學(xué)應(yīng)用基礎(chǔ)研究北京市重點實驗室（北京 100069） 通信作者：武博。E-mail：[email protected]

關(guān)鍵詞：全景拼接；手術(shù)場景；高關(guān)注度區(qū)域；光學(xué)定位；深度信息

分類號：R318.04

出版年·卷·期（頁碼）：2022·41·3（242-248）

摘要：

目的提出一種光學(xué)定位技術(shù)輔助的手術(shù)場景全景拼接算法(panoramic stitching assisted by optical positioning technique, PSAOP)，以提高全景拼接質(zhì)量。方法通過光學(xué)定位系統(tǒng)獲得手術(shù)場景高關(guān)注度區(qū)域中嚴(yán)格匹配的光學(xué)定位點對。結(jié)合SIFT特征匹配點對，建立更準(zhǔn)確的局部單應(yīng)性模型。為了進(jìn)一步減少透視失真，提出基于深度信息優(yōu)化的全局相似變換(depth optimized global similarity transformation algorithm, DOGST)算法來補償攝像機(jī)的運動。結(jié)果對模擬手術(shù)場景進(jìn)行拼接，該算法可以極大提高場景拼接的自然性，獲得了高主觀質(zhì)量。對齊后，視圖之間重疊區(qū)域的結(jié)構(gòu)相似性(structural similarity, SSIM)得分平均高于其他算法。DOGST獲得的全局相似矩陣最小旋轉(zhuǎn)角度小于RANSAC獲得的角度，提高了拼接結(jié)果的自然度。結(jié)論 PSAOP能夠提供更自然的拼接效果，在重疊區(qū)域中沒有明顯的重影，且進(jìn)一步減少了非重疊區(qū)域中的透視失真。未來可與VR(Virtual Reality)結(jié)合使用，為遠(yuǎn)程醫(yī)療提供更身臨其境、更立體直觀的醫(yī)療場景。

Objective To propose a panoramic stitching assisted by optical positioning (PSAOP) technique for surgical scenes for surgical scenes，and to improve the quality of panoramic stitching, Methods The strictly matching optical positioning point pairs in high-attention area of the surgical scenes are obtained by the optical positioning system. Combining SIFT feature matching point pairs, a more accurate local homography model is established. In order to further reduce the perspective distortion, a depth optimized global similarity transformation (DOGST) algorithm is proposed to compensate for the camera motion. Results Experiments on simulated surgery scenes demonstrate that PSAOP can greatly improve the naturalness of the stitching result and obtain high subjective quality. Through subjective observation, the PSAOP can greatly improve the naturalness of stitching. The structural similarity (SSIM) scores of the overlapping region between views after alignment are higher than the other algorithms in average. The minimum rotation angle of global similarity transformation obtained by DOGST is smaller than that obtained by RANSAC, which improves the naturalness of stitching results. Conclusions The PSAOP algorithm provides a more natural stitching with no visible ghosting in the overlapping regions and further reduces the perspective distortion in the non-overlapping regions. Combined with Virtual reality (VR) in the future, it can provide more immersive and intuitive medical situations for telemedicine.

參考文獻(xiàn)：

[1].? Ahrendt A, Schneider H, Kaiser M,et al. Prototypical development of a remote patient monitoring system for efficient treatment of COPD patients[J]. Biomedical Engineering / Biomedizinische Technik, 2013,58(1):4317-4318.

[2].? Ji H, Wang J, Gao J, Liu X. Research on telemedicine technology and implement based on virtual reality[C]//Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC). Xi'an：IEEE,2016.

[3].? Szeliski R. Computer vision: algorithms and applications[M]. New York：Springer-Verlag, Inc, 2010.

[4].? Zhang Y, Lai YK, Zhang F. Content-preserving image stitching with piecewise rectangular boundary constraints[J]. IEEE Transactions on Visualization and Computer Graphics, 2021, ?27（7）：3198-3212.

[5].? 張威. 不同光照條件下的圖像拼接技術(shù)研究[D].沈陽：沈陽工業(yè)大學(xué),2017.

Zhang W. Research on image stitching under different illumination[D]. Shenyang: Shenyang University of Technology,2017.

[6].? Lowe DG. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision,2004,60(2):91-110.

[7].? Gao J, Kim SJ, Brown MS. Constructing image panoramas using dual-homography warping[C]// IEEE Computer Vision and Pattern Recognition (CVPR). Colorado Springs, USA：IEEE，2011.

[8].? Lin W, Liu S, Matsushita Y,et al. Smoothly varying affine stitching[C]//Computer Vision & Pattern Recognition, Colorado Springs, CO, USA：IEEE，2011.

[9].? Zaragoza J, Chin T, Tran Q,et al. As-projective-as-possible image stitching with moving DLT[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014,36(7):1285-1298.

[10]. Chang C, Sato Y, Chuang Y. Shape-preserving half-projective warps for image stitching[C]//IEEE Computer Vision and Pattern Recognition (CVPR). Columbus, OH, USA：IEEE, 2014.

[11]. Zheng J, Wang Y, Wang H, et al. A novel projective-consistent plane based image stitching method[J]. IEEE Transaction on Multimedia，2019,21(10):2561-2575.

[12]. Lin CC, Pankanti SU, Ramamurthy KN,et al. Adaptive as-natural-as-possible image stitching[C]//Computer Vision & Pattern Recognition. Boston, USA：IEEE, 2015.

[13]. Li J, Wang Z, Lai S,et al. Parallax-tolerant image stitching based on robust elastic warping[J]. IEEE Transactions On Multimedia, 2018,20(7):1672-1687.

[14]. Chen YS, Chuang YY. Natural image stitching with the global similarity prior[C]// European Conference on Computer Vision. Amsterdam, Netherlands：Springer International Publishing, 2016.

[15]. Zhang F, Liu F. Parallax-tolerant image stitching[C]// IEEE Computer Vision and Pattern Recognition (CVPR).Columbus, USA：IEEE, 2014.

[16]. Wang Z, Yang Z. Seam elimination based on curvelet for image stitching[J]. Soft Computing, 2019,23(13):5065-5080.

[17]. Lin K, Jiang N, Cheong L,et al. SEAGULL: seam-guided local alignment for parallax-tolerant image stitching[C]// Amsterdam, Netherlands：European Conference on Computer Vision. Springer (ECCV), 2016.

[18]. Hejazifar H, Khotanlou H. Fast and robust seam estimation to seamless image stitching[J]. Signal Image and Video Processing, 2018,12(1):1-9.

[19]. Jiang G, Luo M, Bai K. Optical positioning technology of an assisted puncture robot based on binocular vision[J]. International Journal of Imaging Systems & Technology, 2019, 29(2):180-190.

[20]. Zhou Z, Wu B, Duan J,et al. Optical surgical instrument tracking system based on the principle of stereo vision[J]. Journal of Biomedical Optics, 2017,22(6):1-14.

[21]. Zhang M, Wu B, Ye C,et al. Multiple instruments motion trajectory tracking in optical surgical navigation[J]. Optics Express, 2019,27(11):1-19.

[22]. 梁楠, 賈博奇, 張夢詩,等. 基于改進(jìn)尺度不變特征變換的手術(shù)室多視點圖像拼接算法[J]. 北京生物醫(yī)學(xué)工程, 2018,37(1):9-14.

Zhang MS, Jia BQ, Liang N,et al. Optical tracking algorithm based on motion vectors for multiple surgical instruments[J]. Beijing Biomedical Engineering, 2018,37(4):345-350.

[23]. Fischler MA, Bolles RC. 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.

[24]. Wang Z, Bovik AC, Sheikh HR, et al. Image quality assessment: from error visibility to structural similarity[J]. IEEE Trans Image Process, 2004,13(4):1-13.

服務(wù)與反饋：

【文章下載】【加入收藏】

提示：您還未登錄，請登錄！點此登錄

51黑料吃瓜在线观看,51黑料官网|51黑料捷克街头搭讪_51黑料入口最新视频