임베디드 비전 카메라용 M12 렌즈

용도에 맞는 S-마운트 렌즈를 선택하는 방법

M12 렌즈 선정에는 시야각, 해상도(MTF), 기계적 호환성이라는 세 가지 결정 요소가 있습니다. 이 세 가지를 적절히 선정하면 나머지 사양(F값, IR 필터링, 밀봉)은 사용 환경에 따라 자연스럽게 결정됩니다. 아래 설명에서는 각 매개변수와 임베디드 시스템에서 가장 중요한 상호 절충점에 대해 다룹니다.

시야각 및 초점 거리 고려 사항

• Field of View (FOV) Requirements Focal length sets how much scene the lens captures. Short focal lengths give mobile robots the wide coverage they need for situational awareness; long focal lengths resolve fine detail for inspection at a distance. Wide-angle M12 lenses generally carry more barrel distortion, which you can correct in software or avoid outright with a low distortion design. Telephoto S-mount lenses hold geometric accuracy over a narrow field. Model coverage, distortion included, with our FOV Calculator, then check the result against the lens’s measured distortion data.
• 피사계 심도(DOF) 최적화 피사계 심도란적절하게 선명하게 표현되는 거리 범위를 말합니다. f-값이 높을수록 피사계 심도는 넓어지지만 광량도 줄어들게 되는데, 이는 조명이 변동이 심한 환경에서 작동하는 보드 카메라에 있어 실질적인 제약 요소입니다. 고정 초점 M12 모듈은 재초점 조정을 피하기 위해 넉넉한 피사계 심도에 의존하므로, 작업 거리가 변동하는 경우, 명목상의 거리뿐만 아니라 양쪽 극한 거리에서도 선명도를 확인해야 합니다. 당사의 피사계 심도 계산기는 허용 가능한 흐림 정도를 추정할 때 픽셀 피치를 고려합니다.
• Sensor Format Compatibility The lens image circle must equal or exceed the sensor diagonal, or the corners vignette. A larger-than-needed image circle is usually acceptable for coverage, since the sensor samples the well-corrected center of the field; run the same CRA, clearance, filter-stack, and MTF checks as any pairing. You do pay for size, weight, and glass you never use, so the format-matched lens usually wins on cost. A 1/2.3" optimized S-mount lens works fine on a 1/3" sensor; the reverse produces dark corners. Our Focal Length Calculator maps sensor format to the focal length that produces your target FOV.

광학 성능 사양

• Resolution and MTF Performance The Modulation Transfer Function describes how much contrast the lens preserves at each spatial frequency. It is the standard measure of lens sharpness. Our starting point is roughly 20% to 30% MTF at the sensor’s Nyquist frequency (1 / (2 × pixel pitch)). That band is a Commonlands rule of thumb rather than a published standard, so validate it against the contrast your own task needs. When lens contrast at Nyquist falls well below it, the lens rather than the sensor sets the resolution limit. Over-specifying MTF adds cost without visible benefit once pixel size becomes the limit, so match the lens to the sensor you actually use. Commonlands publishes measured MTF data on request.
• F값과 광량 수집 F값은광량 수집량을 결정합니다. 이미지 평면의 조도는 1/N²에 비례하므로, F1.4 렌즈는 F2.0 렌즈보다 약 두 배 많은 빛을 받아들입니다. 밝은 조리개(F1.4–F2.0)는 저조도 환경에서 신호 대 잡음비를 높여주고, 모션 블러를 줄여주는 짧은 노출 시간을 가능하게 하지만, 일반적으로 화면 가장자리에서 더 많은 수차 현상이 나타나고 피사계 심도가 얕아집니다. 피사체가 충분히 밝고 속도보다 구도나 형태가 더 중요하다면, F2.8 전후의 중간 조리개를 선택하는 것이 종종 더 나은 절충안입니다.
• Chief Ray Angle (CRA) Matching CMOS sensors place microlenses over each photodiode, tuned for a ray incidence angle that increases toward the sensor edge. When the lens CRA diverges too far from the sensor’s microlens design, off-axis light partially misses the photodiodes and produces wavelength-dependent shading: the pink-green corner tint that is difficult to remove in post-processing. Compare the lens CRA plot against the sensor’s published curve point by point across image height rather than as a single number. How much mismatch you can absorb depends on the working F-number and the width of the ray cone, the wavelengths in use, the pixel and filter stack, and the corner shading and MTF loss your pipeline tolerates. Monochrome sensors escape the color shading but still lose light off-angle. Global shutter machine vision sensors such as the OV9281 and AR0234 publish CRA curves in their datasheets.

환경적 및 기계적 요인

• Ingress Protection Requirements An IP67-rated M12 lens seals at the barrel, which makes a window-less build an architecture option: dropping the protective window and its two air-glass surfaces means less stray light and better contrast, but take that option only after validating the full installed assembly against your actual exposure. The rating applies to the tested lens variant in its mounted configuration under the dust and water exposures of IEC 60529, and it does not protect the sensor board or the rest of the camera. For outdoor builds, most teams start their evaluation with the sealed IP67 variant; a washdown jet sits outside what the IP67 immersion test covers, which moves the evaluation to a jet-rated IP6K9K option. Both cases still need the installed assembly validated against the actual exposure, and the housing gets sealed on its own terms; controlled indoor environments can save the cost with standard barrels.
• Thermal Stability Considerations Optical glass has low thermal expansion and low dn/dT, and all-glass M12 lenses are generally rated for −40°C to +85°C operation. A temperature rating is not a focus guarantee: whether focus holds across that range is a property of the assembled lens, holder, and sensor together, and is verified by measuring focus or MTF across the temperatures the camera will see. Hybrid glass-plastic S-mount designs cost less but shift focus with temperature, because optical plastics expand more and their refractive index changes faster with heat. Treat all-glass construction as a first-pass screen for outdoor board cameras facing large temperature swings, then qualify the assembly with an over-temperature measurement; for temperature-controlled indoor systems, hybrid designs are a sound cost optimization.