臺灣博碩士論文加值系統

少子化及高齡化的衝擊下，勞動力缺口不斷擴大，許多產業無法聘用足夠的人力營運；近年來不少企業整合人工智慧(Artificial Intelligence，AI)技術與邊緣運算(Edge Computing)技術，不但減輕了企業對人事安排的負擔，也使從業員工減少部分繁瑣的工作。在過去，營業單位不論是貨品的點交亦或是庫存的清點都需要分配人力進行作業，如今通過人工智慧及邊緣運算技術的結合能夠使其自動進行部分作業，緩解了一部份營業單位的缺工壓力。
通過圖像處理將原始圖像區分前景(物件)與背景後，並將其轉換為二值化(Binary)圖像，最終使用物件標記算法將不同的物件標上不同的編號，我們可以為圖像中的物件進行相對位置分析、物件尺寸分析及計算物件數量，並衍生出清點產品數量等應用；在本論文中，結合鏡頭、顯示器與現場可程式化邏輯陣列(Field Programmable Gate Array，FPGA)裝置去進行圖像處理後為不同的物件進行標籤，在實際驗證中能得到正確的結果。

Under the impact of lower birthrate and aging population, the short of labor become a serious issue. In recent years, many industries have combined artificial intelligence (AI) technology and edge computing technology, which can perform real-time analysis of on-site data. It not only relieves employees of some tedious work, but also enables industries to have more flexibility in personnel arrangements.
After the original image is distinguished between the foreground (object) and the background, the image is converted into a binary image. Then use the object labeling method to label different objects with different numbers. We can use the result to calculate the number of objects, and derive appropriate applications. In this paper, we combine camera, monitor and field programmable gate array (FPGA) device to perform image processing and object labeling. We can obtain accurate results in some practical experiments.

摘 要 VII
ABSTRACT VIII
致謝 IX
目錄 X
圖目錄 XII
表目錄 XIV
第一章 緒論 1
1.1 前言 1
1.2 論文架構 2
第二章 開發平台探討 3
2.1 現場可程式化邏輯陣列介紹[5] 3
2.2 現場可程式化邏輯陣列開發板與選擇[5] 5
2.2.1 Altera DE10-Nano Kit [13] 5
2.2.2 Altera DE0 [16] 5
2.2.3 Xilinx Nexys Video [17] 6
2.2.4 開發平台之評估 7
第三章 圖像處理及連通物件標籤方法探討 9
3.1 連通元件標籤化說明[18] 9
3.2 使用併查集進行連通元件標籤化[24] 9
3.3 使用由粗到細策略進行連通元件標籤化[26] 10
3.4 所提出之使用多次圖像讀取進行連通元件標籤化方法 11
3.5 各種連通元件標籤化方法之比較[24,26] 17
第四章 連通元件標籤化系統 19
4.1 連通元件標籤化系統總覽暨運作流程[15] 19
4.1.1 連通元件標籤化系統之硬體[15] 19
4.1.2 連通元件標籤化系統之軟體[15] 20
4.1.3 連通元件標籤化系統之數據訊號[15] 20
4.2 連通元件標籤化系統各模組說明 21
4.2.1 連通元件標籤化系統之系統流程圖 21
4.2.2 圖像二值化模組之介紹[23] 22
4.2.3 圖像處理模組之介紹[32] 25
4.2.4 連通元件標籤化模組之介紹[23] 29
4.2.5 標籤尺寸統計模組之介紹 31
4.2.6 最終影像輸出模組之介紹[11] 32
4.3 連通元件標籤化系統實作結果 36
4.3.1 圖像二值化模組實作 36
4.3.2 圖像處理模組實作 37
4.3.3 連通元件標籤化模組實作 38
4.3.4 標籤尺寸統計模組實作 39
4.3.5 最終影像輸出模組實作 41
第五章 結論與未來研究方向 42
5.1 結論 42
5.2 未來研究方向 42
參考文獻 43

[1] 行政院。亞洲・矽谷2.0—打造臺灣成為亞洲數位創新的關鍵力量[Online] Available: https://www.ey.gov.tw/Page/5A8A0CB5B41DA11E/c7cdfbb1-13b3-46aa-ac7d-b8518dbe404d
[2] 行政院。加速推動智慧機械產業—使臺灣成為「亞洲高階製造中心」[Online] Available: https://www.ey.gov.tw/Page/5A8A0CB5B41DA11E/1af112a3-80a1-4f45-bc46-f1ca48d16a30
[3] 科技部。智慧終端半導體關鍵技術大突破，邁向下世代半導體新局[Online] Available:https://www.most.gov.tw/folksonomy/detail/1bccc609-2d56-452c-829c-f24d901e4537?l=ch
[4] 威視康。影像辨識實例介紹[Online] Available: https://www.vsk.com.tw/industry-case/87-showcase2/293-vision-sensor-59.html?gsearch=1&moduleId=150&Itemid=481
[5] S. Rai, P. Nath, A. Rupani, S. K. Vishvakarma and A. Kumar, "A Survey of FPGA Logic Cell Designs in the Light of Emerging Technologies," in IEEE Access, vol. 9, pp. 91564-91574, 2021, doi: 10.1109/ACCESS.2021.3092167.
[6] W. Wang, J. Yan, N. Xu, Y. Wang and F. Hsu, "Real-Time High-Quality Stereo Vision System in FPGA," in IEEE Transactions on Circuits and Systems for Video Technology, vol. 25, no. 10, pp. 1696-1708, Oct. 2015, doi: 10.1109/TCSVT.2015.2397196.
[7] W. Xiang et al., "FPGA-Based Two-Dimensional Matched Filter Design for Vein Imaging Systems," in IEEE Journal of Translational Engineering in Health and Medicine, vol. 9, pp. 1-10, 2021, Art no. 1800510, doi: 10.1109/JTEHM.2021.3119886.
[8]S. Li, Y. Luo, K. Sun, N. Yadav and K. K. Choi, "A Novel FPGA Accelerator Design for Real-Time and Ultra-Low Power Deep Convolutional Neural Networks Compared With Titan X GPU," in IEEE Access, vol. 8, pp. 105455-105471, 2020, doi: 10.1109/ACCESS.2020.3000009.
[9] H. Cho, J. Lee and J. Lee, "FARNN: FPGA-GPU Hybrid Acceleration Platform for Recurrent Neural Networks," in IEEE Transactions on Parallel and Distributed Systems, vol. 33, no. 7, pp. 1725-1738, 1 July 2022, doi: 10.1109/TPDS.2021.3124125.
[10] C. Bao, T. Xie, W. Feng, L. Chang and C. Yu, "A Power-Efficient Optimizing Framework FPGA Accelerator Based on Winograd for YOLO," in IEEE Access, vol. 8, pp. 94307-94317, 2020, doi: 10.1109/ACCESS.2020.2995330.
[11] 牟新剛著. (2017, Jan01). 基於FPGA的數位圖像處理原理及應用. 電子工業出版社.
[12] J. Han, D. Moon and M. Meyyappan, "One Time Programmable Antifuse Memory Based on Bulk Junctionless Transistor," in IEEE Electron Device Letters, vol. 39, no. 8, pp. 1156-1158, Aug. 2018, doi: 10.1109/LED.2018.2848975.
[13] Terasic. DE10-Nano Kit. Terasic. Taiwan. [Online] Available: https://www.terasic.com.tw/cgi-bin/page/archive.pl?Language=Taiwan&No=1047
[14] Terasic. LT24 Card. Terasic. Taiwan. [Online] Available:https://www.terasic.com.tw/cgi-bin/page/archive.pl?Language=Taiwan&No=906
[15] Terasic. D8M GPIO User Manual. Terasic. Taiwan. [Online] Available:https://www.terasic.com.tw/cgi-bin/page/archive.pl?Language=Taiwan&CategoryNo=82&No=1013&PartNo=4#contents
[16] Terasic. DE0 User manual. Terasic. Taiwan. [Online] Available: https://www.terasic.com.tw/cgi-bin/page/archive.pl?Language=Taiwan&CategoryNo=173&No=371&PartNo=4#contents
[17] Digilent. Nexys Video Artix-7 FPGA. Digilent. [Online] Available: https://digilent.com/shop/nexys-video-artix-7-fpga-trainer-board-for-multimedia-applications/
[18] H. Samet and M. Tamminen, "Efficient component labeling of images of arbitrary dimension represented by linear bintrees," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 10, no. 4, pp. 579-586, July 1988, doi: 10.1109/34.3918.
[19] R. Cucchiara, C. Grana, A. Prati and R. Vezzani, "Computer vision techniques for PDA accessibility of in-house video surveillance", Proc. 1st ACM SIGMM Int. Workshop Video Surveillance, pp. 87-97, Nov. 2003. A. A. Galadima, "Arduino as a learning tool," 2014 11th International Conference on Electronics, Computer and Computation (ICECCO), Abuja, 2014, pp. 1-4.
[20] H. V. Pham, B. Bhaduri, K. Tangella, C. Best-Popescu and G. Popescu, "Real time blood testing using quantitative phase imaging", Plos One, vol. 8, no. 2, 2013. G. K. Pandey and P. Chinnamuthu, "IOT Based Patient Monitoring System utilizing Raspberry Pi and Web-Page," 2018 3rd IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT), Bangalore, India, 2018, pp. 2140-2144.
[21] T. Lelore and F. Bouchara, "FAIR: A fast algorithm for document image restoration", IEEE Trans. Pattern Anal. Mach. Intell., vol. 35, no. 8, pp. 2039-2048, Aug. 2013.
[22] S. Byna et al., "Detecting atmospheric rivers in large climate datasets", Proc. 2nd Int. Workshop Petascal Data Analytics Challenges Opportunities, pp. 7-14, 2011.
[23] J. M. Ponce, A. Aquino, B. Millan and J. M. Andújar, "Automatic Counting and Individual Size and Mass Estimation of Olive-Fruits Through Computer Vision Techniques," in IEEE Access, vol. 7, pp. 59451-59465, 2019, doi: 10.1109/ACCESS.2019.2915169.
[24] J. Xu et al., "Asynchronous and Load-Balanced Union-Find for Distributed and Parallel Scientific Data Visualization and Analysis," in IEEE Transactions on Visualization and Computer Graphics, vol. 27, no. 6, pp. 2808-2820, June 2021, doi: 10.1109/TVCG.2021.3074584.
[25] 吳燦銘著. (2020, Dec30). 圖解資料結構：使用C++. 博碩出版社.
[26] J. Chen, K. Nonaka, H. Sankoh, R. Watanabe, H. Sabirin and S. Naito, "Efficient Parallel Connected Component Labeling With a Coarse-to-Fine Strategy," in IEEE Access, vol. 6, pp. 55731-55740, 2018, doi: 10.1109/ACCESS.2018.2872452.
[27] K. A. Hawick, A. Leist and D. P. Playne, "Parallel graph component labelling with GPUs and CUDA", Parallel Comput., vol. 36, no. 12, pp. 655-678, Dec. 2010.
[28] G. Viais de Brito Silva and F. César Flores, "Rot Corn Grain Classication by Color and Texture Analysis," in IEEE Latin America Transactions, vol. 20, no. 2, pp. 208-214, Feb. 2022, doi: 10.1109/TLA.2022.9661459.
[29] F. Bolelli, S. Allegretti, L. Baraldi and C. Grana, "Spaghetti Labeling: Directed Acyclic Graphs for Block-Based Connected Components Labeling," in IEEE Transactions on Image Processing, vol. 29, pp. 1999-2012, 2020, doi: 10.1109/TIP.2019.2946979.
[30] H. Zhou, R. Dou, L. Cheng, J. Liu and N. Wu, "A Provisional Labels-Reduced, Real-Time Connected Component Labeling Algorithm for Edge Hardware," in IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 69, no. 6, pp. 2997-3001, June 2022, doi: 10.1109/TCSII.2022.3152783.
[31] Altera. (2016, May). Embedded Memory (RAM: 1-PORT, RAM: 2-PORT, ROM: 1-PORT, and ROM: 2-PORT) User Guide. Altera. America. [Online] Available: https://people.ece.cornell.edu/land/courses/ece5760/DE1_SOC/ug_ram_rom.pdf
[32] R. C. Gonzalez and R. E. Woods, Digital image processing, 4^thedition, Pearson Publishers, Inc. 2018.
[33] Altera. Signal Tap Logic Analyzer: Introduction & Getting Started. Altera. [Online] Available: https://learning.intel.com/developer/learn/course/external/view/elearning/203/signal-tap-logic-analyzer-introduction-getting-started
[34] B. C. C. Leme, L. F. Almeida, J. W. P. Bizarria, F. C. P. Bizarria, A. M. S. Soares and M. A. C. Ramos, "Development of a low-cost tool for semi-automatic classification and counting of particles in industrial oils," 2017 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM), 2017, pp. 1925-1929, doi: 10.1109/IEEM.2017.8290227.