2015考研英语阅读英文原刊《经济学人》：识别人们的行踪

　　2015考研英语复习正是强化复习阶段，考研英语阅读在考研英语中占了40分，所以考研英语阅读是英语科目中重要的一项。新东方名师范猛老师曾建议过考研生需要坚持每天泛读10-15分钟的英文原刊。强烈推荐了杂志《经济学人》杂志中的文章也是考研英语的主要材料来源.希望考研考生认真阅读，快速提高考研英语阅读水平。

　　Biometrics

　　计量测定学

　　Clocking people's clocks

　　识别人们的行踪

　　Facial-recognition systems are getting better

　　面部识别系统愈发完善

　　WITH some pride, the FBI trumpeted the news last month that thanks to the agency's facial-recognition system Neil Stammer, wanted for sexual assault and kidnapping, had beenapprehended in Nepal after being on the run for 14 years. The truth was slightly moreprosaic. A State Department official had used the FBI's “Wanted” posters in a test for passportfraud. The system then matched Mr Stammer's face with an American calling himself KevinHodges who regularly visited the US embassy in Kathmandu to renew his visa. Still, MrStammer's arrest illuminates the growing importance of facial-recognition technology.

　　上个月，在FBI面部识别系统的帮助下，因性侵犯和绑架而被通缉长达14年的Neil Stammer在尼泊尔落入法网。FBI对此大肆宣传，还带着点小骄傲呢~ 经过可能较为平淡。一名国务院官员使用FBI的通缉海报来测试护照欺诈行为。然后系统识别出，Stammer的面部同一名自称Kevin Hodges的美国人相符。后者经常到美国驻加德满都大使馆去更新他的签证。不过，Stammer的落网，说明面部识别技术日益重要。

　　The two main techniques used to recognise faces electronically are principal-componentanalysis (PCA) and linear-discriminant analysis (LDA). Both compare a picture of someone'sphizog with a reference image taken in a controlled environment. Passport photos andmugshots, then, are about as ideal as it gets.

　　电子识别人脸的两种技术分别为主成分分析(PCA)和线性判别分析(LDA)。两种方法都是通过在特定环境下比较某人的面部和参照图片。碰巧，护照照片和疑犯照片是最理想化的。

　　Basic PCA and LDA are good for skin colour, hair colour and the like. Advanced systems, suchas that used with British biometric passports, may look at cheek bones, the bridge of the nose,jaw lines and eyes.

　　基本的PCA和LDA擅长处理肤色、发色等。先进系统，比如英国计量生物学护照所采用的，可能会记录面颊骨骼，鼻梁，下颌轮廓以及眼睛。

　　All of which is fine when someone is sitting or standing in front of a camera, but is less useful inthe world beyond the studio. That requires a technique called Elastic Bunch Graph Matching(EBGM), which tries to create a three-dimensional (3D) model from two-dimensional images.This model can, thereafter, be used to match any subsequent image, or part thereof.

　　在人正坐在或者正立在摄像头前的时候，所有这些标准都好办，但是在摄影棚外就不那么有用了。这要求一种名为弹性束图匹配(EBGM)的技术，该技术力求通过2D图像创建一个3D模型。因此这个模型就能用于匹配任何相关图片，或其中的部分。

　　EBGM considers the head as a union of two ellipsoids: one whose main axis is vertical, andruns from forehead to chin; the other whose main axis is horizontal, and runs from tip of thenose to the back of the cranium. This basic scheme is overlaid with “fiducial” points which act asanchors for the modelling. These can be as few as half a dozen (the pupils of the eyes, thecorners of the mouth, and so on), or as many, in one system, as 40,000.

　　EBGM将头部视为两个椭圆的结合体。一个的主轴是垂直的，从前额到下巴;另一个的主轴是水平的，从鼻尖到头盖骨后侧。基本方案是通过确定基点位置，定位模型尺寸。基点少则6个(瞳孔，嘴角，等等)，多则可能在一个系统中达到40000个。

　　EBGM allows the construction of a three-dimensional representation of a face from poorly litimages taken at odd angles, such as a closed-circuit television camera might provide. Once itrecognises enough fiducial points it can work out what aspect of a face it is viewing. It thenextrapolates the expected positions of other fiducial points. As more data come in from thecamera, the model's shape is updated. Given enough horsepower, says a British official, such asystem can build a model from as few as 80 pixels located between a subject's eyes—and onlytwo images are needed for a 3D reconstruction.

　　EBGM可以将小角度拍摄的暗光照片，比如闭路电视所提供的，塑造为三维模型。一旦它捕捉到了足够的基点，系统就能计算出它所观测的是面部的那个位置。然后系统推断出其他基点的预期位置。随着摄像头提供的数据增多，模型的形状不断更新。如果有足够动力，比如英国官方要求，只需要两张照片，像素少到在人物的眼睛中只有80个像素，这个系统就能偶建立出3D复原模型。

　　Governments are not the only ones interested. Earlier this year, Facebook's DeepFace systemwas asked whether thousands of pairs of photos were of the same person. It answeredcorrectly 97.25% of the time, a shade behind humans at 97.53%. Although DeepFace is only aresearch project, and is aided by the fact that many Facebook photos are tagged with thenames of people in the images, which lets the system learn those faces in different poses andlighting conditions, it is still an impressive feat.

　　并不只有政府对此感兴趣。今年早些时候，有人拿脸谱的“深脸”系统测试了成千上万张照片，看他们是否是同一个人。其正确率为97.25%，在人背后后有阴影的情况下识别率为97.53%。尽管深脸只是研究项目，并且脸谱上的许多照片都是贴着人的姓名这个情况给出了一定帮助，使得系统能够辨别出不同姿势、不同采光下的人脸，但是这个成果仍旧令人瞩目。

　　As DeepFace shows, access to an accurate gallery of images is crucial. Passport photos, orthose on national identity cards, can act as such galleries, as they can be rendered by EBGMinto usable 3D models. Add in the increasing ubiquity of closed-circuit television, and the ideathat anyone will be able hide for long in Nepal, or anywhere else, looks quaint.

　　正如深脸所显示的，想精确识别大批量图案依然困难重重。护照照片，或者那些用在国民身份证上的，可以作为这种相片管理系统，因为他们可以可以由EBGM转述为实用3D模型。再考虑到闭路电视的普及，那么任何一个人，想要在尼泊尔或者其他任何地方藏起来的这种想法都会变得不切实际。

【英语阅读资料】这里有↑↑↑