本研究聚焦於監控設備監控模組中日益凸顯的「機構性暗角」問題，旨在提出一套從成因解析到量產導入的系統化改善流程，以提升影像均勻性與產品品質。隨著高解析度與廣視角需求日益增長，鏡頭與機構結合時之遮蔽干涉現象，已對使用者之影像體驗與產品競爭力構成挑戰。
研究首先以量產型監控設備模組為樣本，採用相對照度量測、光路追蹤模擬與實機拍攝三管齊下的方式，對暗角成因進行深入剖析。經分析歸納，主要問題源自：
1.鏡頭模組與導引結構之公差配合間隙不足。
2.鏡頭與印刷電路基板間之定位偏移。
3.鏡頭膠圈之遮蔽設計不良。
針對上述三大關鍵因素，提出以下結構優化策略：重新定義公差配合範圍並採漸進式干涉設計、在印刷電路板上增設定位導柱以確保鏡頭中心對位、以及依據光路模擬結果優化膠圈形狀與高度。
優化方案分別應用於兩款樣機進行實際測試，並於一百台樣機中完成量化評估。結果顯示，優化後影像四個角落之相對灰階值比值平均數，已由原本的28%提升至93%；其灰階值比值之差異振幅亦控制在±1.2%以內，充分驗證本方案具備良好之有效性與一致性。
本研究除建構了一條清晰可行的改良流程外，亦證實所提方法具備高度製造穩定性與設計參考價值。建議未來於智慧監控模組開發與設計準則中納入本成果，以全面提升影像品質與量產可靠性。

This study addresses the increasingly prominent issue of “mechanical vignetting” in smart doorbell camera modules by developing a systematic improvement workflow—from root-cause analysis through pilot verification to mass-production implementation—to enhance image uniformity and overall product quality. As demand for higher resolution and wider field of view continues to grow, shading and occlusion effects arising during the integration of lens and mechanical components have become a critical challenge to user experience and market competitiveness. First, a mass-production doorbell module was selected as the test vehicle. Three complementary approaches—relative luminance measurement, optical ray-trace simulation, and empirical imaging tests—were employed to dissect the causes of vignetting. Analysis revealed three principal contributors: (1) insufficient tolerance clearance between the lens module and its guide structure; (2) misalignment between the lens assembly and the printed circuit board (PCB); and (3) suboptimal design of the lens retaining O-ring that exacerbates shading. To address these issues, we propose three structural optimizations: redefining tolerance ranges with a graduated interference fit; adding alignment pillars on the PCB to constrain lens centering; and refining the O-ring profile and height based on simulation insights. These optimizations were implemented on two prototype variants and validated across 100 units. Results demonstrate that the average corner-to-center luminance ratio improved from 28 % to 92.95 %, with luminance variation maintained within ±1.2 %, confirming both effectiveness and manufacturing consistency. Beyond establishing a clear, repeatable improvement process, this work verifies the proposed methods’ high reproducibility and design value. It is recommended that future smart surveillance module development guidelines incorporate these findings to ensure superior image quality and production reliability.
Keywords: smart surveillance module, mechanical vignetting, relative luminance, tolerance design, lens alignment, mass-production validation

[1] RED, “Understanding Lens Vignetting,” 2025. https://www.red.com/red-101/lens-vignetting
[2] Wikipedia, “Vignetting,” 2025. https://en.wikipedia.org/wiki/Vignetting
[3] Carr Lane, “Locating & Clamping Principles (3-2-1),” 2025. https://www.carrlane.com/engineering-resources/fixture-design-principles/locating-clamping-principles
[4] FARO, “GD&T in Precision Engineering: Using Slots in Precision Location,” 2023. https://www.faro.com/en/Resource-Library/Article/gd-t-in-precision-engineering-using-slots-in-precision-location
[5] GD&T Basics, “True Position – Position Tolerance,” 2025. https://www.gdandtbasics.com/true-position/
[6] Machine Design, “Ensuring Parts Fit: Using Bonus Tolerance,” 2025. https://www.machinedesign.com/cad-cam-software/article/55281116/ensuring-parts-fit-using-bonus-tolerance-to-your-advantage
[7] Fictiv, “Tolerance Stack-Ups 501: Beyond Mechanical Fit,” 2021. https://www.fictiv.com/articles/tolerance-stack-ups-501-going-beyond-mechanical-fit-and-into-predictive-design
[8] McGill University, “Principles of Tolerancing,” 2025. https://www.mcgill.ca/engineeringdesign/step-step-design-process/basics-graphics-communication/principles-tolerancing
[9] NIST/SEMATECH, “2.4 Gauge R&R studies,” e-Handbook. https://www.itl.nist.gov/div898/handbook/mpc/section4/mpc4.htm
[10] MoreSteam, “Measurement System Analysis (MSA),” 2025. https://www.moresteam.com/toolbox/measurement-system-analysis
[11] NIST/SEMATECH, “6.1.6 What is Process Capability?,” e-Handbook. https://www.itl.nist.gov/div898/handbook/pmc/section1/pmc16.htm
[12] Minitab Blog, “Process Capability Statistics: Cpk vs. Ppk,” 2016. https://blog.minitab.com/en/process-capability-statistics-cpk-vs-ppk
[13] 1factory, “Guide: Process Capability,” 2025. https://www.1factory.com/quality-academy/guide-process-capability.html
[14] Mercedes-Benz, “Supplier Quality Assurance Manual,” 2023. https://docmaster.supplier.mercedes-benz.com/DMPublic/en/doc/ALD00001035.2023-04.EN.pdf
[15] ASQ, “What is a Fishbone (Ishikawa) Diagram?,” 2025. https://asq.org/quality-resources/fishbone
[16] A. Kumah, “Cause-and-Effect (Fishbone) Diagram: A Tool for Root Cause Analysis,” J. Multidiscip. Healthcare, 2024. https://pmc.ncbi.nlm.nih.gov/articles/PMC11077513/
[17] NASA GSFC, “Guideline for FMECA (GSFC-HDBK-8004),” 2024. https://standards.nasa.gov/standard/GSFC/GSFC-HDBK-8004
[18] IEC, “Ingress Protection (IP) ratings,” 2025. https://www.iec.ch/ip-ratings
[19] Intertek, “IP Testing per IEC 60529,” 2025. https://www.intertek.com/lighting/performance/ingress-protection/
[20] US Patent, “Optical cap component,” US 11,092,769 B2, 2021. https://patents.google.com/patent/US11092769B2