本論文提出一個影像擷取與重疊的機制，此機制分別在雲端伺服器做物件的擷取，且在邊緣伺服器做畫面的重疊，我們設計一個通訊協定讓兩台伺服器可以做物件的傳送與接收並實作影像擷取與重疊的機制。為了分析影像擷取與重疊所需的網路傳送時間，我們比較FPS (frames per second)在雲端伺服器做影像擷取與在邊緣伺服器做影像重疊的三種情境，第一個情境是當雲端伺服器的FPS大於邊緣伺服器的FPS時，前者只會擷取與後者使用的FPS相同個數的物件以避免不必要的擷取與傳送，第二個情境是當雲端伺服器的FPS等於邊緣伺服器的FPS時，前者會根據自己的FPS依序對每個畫面做擷取，第三個情境是當雲端伺服器的FPS小於邊緣伺服器的FPS時，前者會擷取與自己FPS相同個數的物件並傳送給後者。我們使用數值分析來比較三種FPS的情境下影像擷取與重疊的計算時間與傳送物件的網路時間，在分析計算時間時，我們在雲端伺服器使用矩陣相乘的個數來增加計算的負載，而在分析網路時間時，我們提高背景資料流的傳輸率來增加網路的負擔，從實作與數值分析的結果中，我們證明所提出的影像擷取與重疊的機制在使用邊緣伺服器來分擔雲端伺服器的負載確實可減少影像擷取與重疊的計算時間，且降低兩台伺服器間網路傳送時間。

In this thesis, we propose an image extraction and overlaying mechanism. In this mechanism, a cloud server will do the extraction, while an edge server will do the overlay. For the purpose of inter-communications between the two servers, we design a communication protocol for the servers to send and receive the extracted objects. We then implement the proposed image extraction and overlaying mechanism. To analyze the computation and network transmission time for object extraction and overlay, we design three FPS (frames per second) scenarios. The first scenario suits when the FPS of a cloud server is greater than that of an edge server. In this case, to avoid unnecessary extraction and network transmission, the cloud server will extract the objects per second which is the same as the FPS used by the edge server. The second scenario fits when the FPS of a cloud server is equal to that of an edge server. In this case, a cloud server will just extract the objects one by one from the consecutive frames. The third scenario fits when the FPS of a cloud server is smaller than that of an edge server. In this case, the cloud server will extract the objects per second, which is the same as its own FPS. In addition to the measurements in network transmission time, we use numerical simulation to compare the computation time by increasing the number of matrix multiplications on the cloud server. From the measurements and the simulation results, we validate that the proposed mechanism can indeed reduce the computation time of image extraction and overlay, while significantly reduce the network transmission time between the two servers.

論文審定書 i
致謝 ii
摘要 iii
Abstract iv
目錄 v
圖表目錄 vii
第一章　導論 1
1.1 研究動機 1
1.2 研究方法 2
1.3 章節介紹 3
第二章　MEC網路影音串流的影像擷取與重疊 4
2.1 MEC網路的架構與應用 4
2.1.1 服務轉移與功能分割 4
2.1.2 網路計算的種類 6
2.1.3 行動邊緣計算的優點 7
2.1.3.1 計算負載的分享 7
2.1.3.2 網路延遲的降低 8
2.2 使用RTMP協定傳送影音串流 9
2.3 兩個影像的重疊 11
2.3.1 物件的擷取 11
2.3.1.1 灰階的色值轉換 11
2.3.1.2 灰階的邊緣偵測 12
2.3.1.3 灰階的遮罩處理 13
2.3.2 物件的重疊 14
2.4 相關研究 15
第三章　影像擷取與重疊的機制 21
3.1 影像擷取與重疊架構 21
3.2 物件擷取與傳送 24
3.2.1 畫面張數的訊息的TCP Header 24
3.2.2 在雲端伺服器做擷取物件的切割與傳送 25
3.3 物件接收與重疊 28
3.4 節點的計算與網路傳送時間 32
3.4.1 影像擷取與重疊的計算時間 33
3.4.2 網路傳送的時間 34
3.4.3 三種情境的計算與網路傳送時間 36
第四章　實作與結果分析 40
4.1 影像擷取與重疊的實作 40
4.1.1 Cloud Server的擷取與傳送的虛擬碼 43
4.1.2 Edge Server的接收與重疊的虛擬碼 48
4.1.3 量測結果與分析 53
4.2 分析計算時間與網路時間 55
4.2.1 計算時間 55
4.2.2 網路時間 56
第五章　結論與未來工作 59
5.1 結論 59
5.2 遭遇的困難 60
5.3 未來工作 60
Reference 61
Acronyms 66
Index 68

[1] A. Elgazar and K. Harras, “Teddybear: Enabling Efficient Seamless Container Migration in User-Owned Edge Platforms,” 2019 IEEE International Conference on Cloud Computing Technology and Science (CloudCom), Sydney, Australia, pp. 70-77, Dec. 11-13, 2019.
[2] D. Baburao, T. Pavankumar, and C.S.R. Prabhu, “Survey on Service Migration, load optimization and Load Balancing in Fog Computing Environment,” 2019 IEEE 5th International Conference for Convergence in Technology (I2CT), Bombay, India, pp. 1-5, Mar. 29-31, 2019.
[3] A. Randazzo and I. Tinnirello, “Kata Containers: An Emerging Architecture for Enabling MEC Services in Fast and Secure Way,” 2019 Sixth International Conference on Internet of Things: Systems, Management and Security (IOTSMS), Granada, Spain, pp. 209-214, Oct. 22-25, 2019.
[4] A. Marotta, D. Cassioli, K. Kondepu, C. Antonelli, and L. Valcarenghi, “Efficient Management of Flexible Functional Split through Software Defined 5G Converged Access,” 2018 IEEE International Conference on Communications (ICC), Kansas City, MO, USA, pp. 1-6, May 20-24, 2018.
[5] J. Jeon and J. Kim, “Privacy-Sensitive Parallel Split Learning,” 2020 International Conference on Information Networking (ICOIN), Barcelona, Spain, pp. 7-9, Jan. 1-3, 2020.
[6] A. Sriram, M. Masoudi, A. Alabbasi, and C. Cavdar, “Joint Functional Splitting and Content Placement for Green Hybrid CRAN,” 2019 IEEE 30th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC), Istanbul, Turkey, pp. 1-7, Sept. 8-11, 2019.
[7] K. Dolui and S. K. Datta, “Comparison of edge computing implementations: Fog computing, cloudlet and mobile edge computing,” 2017 Global Internet of Things Summit (GIoTS), Geneva, Switzerland, pp. 1-6, June 6-9, 2017.
[8] M. Marjanović, A. Antonić, and I. P. Žarko, “Edge Computing Architecture for Mobile Crowdsensing,” IEEE Access, Vol. 6, USA, pp. 10662-10674, Jan. 30, 2018.
[9] J. Balcerek, A. Konieczka, T. Marciniak, A. Dąbrowski, K. Maćkowiak, and K. Piniarski, “Video processing approach for supporting pedestrians in vehicle detection,” 2014 Signal Processing: Algorithms, Architectures, Arrangements, and Applications (SPA), Poznan, Poland, pp. 100-103, Sept. 22-24, 2014.
[10] A. Karakottas, A. Papachristou, A. Doumanoqlou, N. Zioulis, D. Zarpalas, and P. Daras, “Augmented VR,” 2018 IEEE Conference on Virtual Reality and 3D User Interfaces (VR), Reutlingen, Germany, pp. 831, Mar. 18-22, 2018.
[11] D. D. Dominicis and D. Accardo, “Software and Sensor Issues for Autonomous Systems based on Machine Learning Solutions,” 2020 IEEE 7th International Workshop on Metrology for AeroSpace (MetroAeroSpace), Pisa, Italy, pp. 545-549, June 22-24, 2020.
[12] H. J. Jeong, “Lightweight Offloading System for Mobile Edge Computing,” IEEE Photonics Technology Letters,” 2019 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops), Kyoto, Japan, pp. 307-310, Mar. 2019.
[13] C. H. Hung, Y. C. Hsieh, and L. C. Wang, “Control plane latency reduction for service chaining in mobile edge computing system,” 2017 13th International Conference on Network and Service Management (CNSM), Kyoto, Japan, pp. 1-50, Nov. 26-30, 2017.
[14] T. Böttger, F. Cuadrado, G. Tyson, I. Castro, and S. Uhlig, “Open Connect Everywhere: A Glimpse at the Internet Ecosystem through the Lens of the Netflix CDN,” ACM SIGCOMM Computer Communication Review, Vol. 48 Issue, USA, pp. 28-34, Jan. 2018.
[15] Adobe Systems Inc., “RTMP Specification 1.0”, June 2009.
[16] Adobe Systems Inc.,”Video File Format Specification Version 10”, Nov. 2008.
[17] A. Aloman, A.I. Ispas, P. Ciotirnae, R. Sanchez-Iborra, and M.D. Cano, “Performance Evaluation of Video Streaming Using MPEG DASH, RTSP, and RTMP in Mobile Networks,” 2015 8th IFIP Wireless and Mobile Networking Conference (WMNC), Munich, Germany, pp. 144-151, Oct. 5-7, 2015.
[18] R. C. Gonzalez and R. R. Woods,” Digital Image Processing 3rd Edition (State of New Jersey, Pearson, 2007)”, July 2007.
[19] A Akter, A Islam, and S Y Shin, “Mobile Edge Computing based Mixed Reality Application for the Assistance of Blind and Visually Impaired People,” 2019 7th International Conference on Information and Communication Technology (ICoICT), Kuala Lumpur, Malaysia, pp. 1-5, July 24-26, 2019.
[20] M. Pore, V. Chakati, A. Banerjee, and S. K. S. Gupta, “Poster Abstract: ContextMete: Context Optimized Peer Offload Architecture for Distributed Mobile Applications,” 2016 IEEE/ACM Symposium on Edge Computing (SEC), Washington, DC, USA, pp. 105-106, Oct. 27-28, 2016.
[21] J. Yu, A. Vandanapu, C. Qu, S. Wang, and P. Calyam, “Energy-aware Dynamic Computation Offloading for Video Analytics in Multi-UAV Systems,” 2020 International Conference on Computing, Networking and Communications (ICNC), Big Island, HI, USA, pp. 641-647, Feb. 17-20, 2020.
[22] E. Baccour, A. Erbad, A. Mohamed, and M. Guizani, “CE-D2D: Dual Framework Chunks Caching and offloading in Collaborative Edge networks with D2D communication,” 2019 15th International Wireless Communications & Mobile Computing Conference (IWCMC), Tangier, Morocco, pp. 1550-1556, June 24-28, 2019.
[23] S. Takagi, J. Kaneda, S. Arakawa, and M. Murata, “An Improvement of Service Qualities by Edge Computing in Network-oriented Mixed Reality Application,” 2019 6th International Conference on Control, Decision and Information Technologies (CoDIT), Paris, France, pp. 773-778, Apr. 23-26, 2019.
[24] R. R Ramisetty, C. Qu, R. Aktar, S. Wang, P. Calyam, and K. Palaniappan, “Dynamic Computation Off-loading and Control based on Occlusion Detection in Drone Video Analytics,” 2020 Proceedings of the 21st International Conference on Distributed Computing and Networking (ICDCN), New York, USA, pp. 1-10, Jan. 2020.
[25] D. Chemodanov, C. Qu, O. Opeoluwa, S. Wang, and P. Calyam, “Policy-Based Function-Centric Computation Offloading for Real-Time Drone Video Analytics,” 2019 IEEE International Symposium on Local and Metropolitan Area Networks (LANMAN), Paris, France, pp. 1-6, July 1-3, 2019.
[26] B Dab, N Aitsaadi, and R Langar, “Q-Learning Algorithm for Joint Computation Offloading and Resource Allocation in Edge Cloud,” 2019 IFIP/IEEE Symposium on Integrated Network and Service Management (IM) , Arlington, VA, USA, pp. 45-52, Apr. 8-12, 2019.
[27] M. Vassell, O. Apperson, P. Calyam, J. Gillis, and S. Ahmad, “Intelligent Dashboard for augmented reality based incident command response co-ordination,” 2016 13th IEEE Annual Consumer Communications & Networking Conference (CCNC), Las Vegas, NV, USA, pp. 976-979, Jan. 9-12, 2016.
[28] A. V. Kempen, T. Crivat, B. Trubert, D. Roy, and G. Pierre, “MEC-ConPaaS: An Experimental Single-Board Based Mobile Edge Cloud,” 2017 5th IEEE International Conference on Mobile Cloud Computing, Services, and Engineering (MobileCloud), San Francisco, CA, USA, pp. 17-24, Apr. 6-8, 2017.