Photography
摄影
A frame camera shooting 20 images a second supplies data at 50-millisecond intervals. Event cameras, unconstrained by such a fixed timetable, can respond in microseconds. Davide Scaramuzza, director of the Robotics and Perception Group at the University of Zurich, has, like Dr de Croon, been working with event cameras mounted on drones. Collision-detection cameras usually take 50-200 milliseconds to react. His team used an event camera to cut this below four milliseconds, enabling faster manoeuvring whilst avoiding obstacles.
一台每秒拍摄20幅图像的框幅相机以50毫秒的间隔提供数据。事件相机不受这种固定时间的限制，可以在微秒内做出反应。和德克鲁恩一样，苏黎世大学机器人与感知小组的负责人达维德•斯卡拉穆扎也一直在研究安装在无人机上的事件相机。防撞摄像头通常需要50~200毫秒来进行反应。他的团队通过使用事件相机将这个时间缩短至不到4毫秒，从而加快了避开障碍时的机动速度。
Response time is equally important for self-driving cars. Dr Scaramuzza's team are working on event cameras for this application with what he terms a “top-tier” company in the automotive sector. Event cameras bring benefits besides speed. Frame-based moving-picture cameras capture redundant information, such as areas of unchanging blue sky. To avoid wasting disk space and bandwidth, images taken this way are often compressed by special software, such as MPEG, for storage or transmission. But event-camera images do not require compression. According to Dr Scaramuzza, they are about 40 times as efficient in this respect as frame cameras.
对于无人驾驶汽车来说，反应时间同样重要。斯卡拉穆扎的团队正在与他称之为汽车行业的“顶级”公司进行合作，共同研究事件相机在该领域的应用。事件相机不仅仅只是速度快。基于框幅的摄像机会捕捉到冗余信息，比如一成不变的蓝天。为了避免浪费磁盘空间和带宽，以这种方式拍摄的图像通常由特殊软件进行压缩（如MPEG），以便进行储存或传输，但事件相机拍摄的照片却无需压缩。据斯卡拉穆扎博士说，它们的效率大概是框幅相机的40倍。
On top of that, event cameras are practically immune to motion blur. Capturing images of speeding bullets is no problem, and they can film the hare and the tortoise at the same time with equal clarity. They could thus solve the perennial problem of extracting unblurred stills from closed-circuit television feeds. Yet another advantage of event cameras is that they cope easily with bright and dim lighting in a single image. A frame camera on a self-driving car might miss a pedestrian in a shadow beside a well-lit street-with disastrous consequences. An event camera will catch the slightest movement, even in deep shade.
除此之外，事件相机几乎不受运动模糊的影响。它不仅能轻而易举地拍摄高速飞行的子弹，还能以同样的清晰度在同一时间拍摄兔子和乌龟。受益于这个优点，他们可以解决一个悬而未决的难题：从闭路电视视频流中提取不模糊的定格画面。事件相机的另一个优点体现在它们可以轻松处理同一张图像中光线的明暗。无人驾驶汽车上的框幅相机可能会在明亮街道旁的阴影中难以识别行人，造成灾难性后果。然而，即使是在浓阴下，事件相机也能捕捉到最细微的运动。
These benefits make event cameras attractive, but they are still fairly new and unknown. The first commercial examples, which appeared in 2014, had a resolution of just 100x100 pixels. Versions branded as “high resolution” appeared in 2019, but all things are relative. The 640x480 pixel resolution of these machines matches that of Apple’s QuickTake camera of 1994. Something closer to real high resolution may be coming soon, though. Both Samsung and Sony are working on event cameras for the mass market. Sony's will have a recording field measuring 1,280x720 pixels.
虽然这些优点使得事件相机颇受瞩目，但它们还处于全新和未知的领域。2004年面世的首批商用产品的分辨率仅为100X100像素，而五年后，以“高分辨率”著称的相机便应运而生。但一切都是相对的——这些640×480像素的相机只能和苹果1994年上架的QuickTake 相匹敌。不过，接近真正高分辨率的产品也许不久就会问世。三星和索尼都在研究面向大众市场的事件相机。索尼的相机将有一个1280×720像素的记录视野。
来源：经济学人
检索：倪珂
翻译：谢鑫飞
审校、上传：宋仪欢
