Dan Berkenstock: The world is one big dataset. Now, how to photograph it ...

61,716 views ・ 2014-02-04

TED


请双击下面的英文字幕来播放视频。

翻译人员: RUI FAN 校对人员: Shiwen He
00:12
Five years ago, I was a Ph.D. student
0
12498
2599
五年前,我是个博士生
00:15
living two lives.
1
15097
1644
过着两种生活
00:16
In one, I used NASA supercomputers
2
16741
2316
在一种生活里 我用NASA(美国国家航空航天局)的超级电脑
00:19
to design next-generation spacecraft,
3
19057
2728
设计新一代的航天器
00:21
and in the other I was a data scientist
4
21785
2579
在另一种生活里,我是一个数据科学家
00:24
looking for potential smugglers
5
24364
2419
寻找那些敏感核技术的
00:26
of sensitive nuclear technologies.
6
26783
3239
潜在走私者
00:30
As a data scientist, I did a lot of analyses,
7
30022
2859
作为一个数据科学家,我做了很多的分析
00:32
mostly of facilities,
8
32881
1467
大部分是对设施的分析
00:34
industrial facilities around the world.
9
34348
2271
世界各地的工业设施
00:36
And I was always looking for a better canvas
10
36619
2688
而我一直在寻找一个可以更好的框架
00:39
to tie these all together.
11
39307
2066
可以将这些都联结起来
00:41
And one day, I was thinking about how
12
41373
1866
有一天,我正在思考
00:43
all data has a location,
13
43239
2124
所有的数据是如何定位的
00:45
and I realized that the answer
14
45363
1733
然后我意识到
00:47
had been staring me in the face.
15
47096
1888
答案一直就在我的眼前
00:48
Although I was a satellite engineer,
16
48984
2638
虽然我是一个卫星工程师
00:51
I hadn't thought about using satellite imagery
17
51622
3171
我却从来没有想过将卫星图像
00:54
in my work.
18
54793
1566
运用到我的工作中
00:56
Now, like most of us, I'd been online,
19
56359
1947
现在,我会像大多数人一样
00:58
I'd see my house, so I thought,
20
58306
1782
上网看看自己的房子,于是我想
01:00
I'll hop in there and I'll start looking up
21
60088
2298
我也可以开始研究研究
01:02
some of these facilities.
22
62386
1417
其中的一些设施
01:03
And what I found really surprised me.
23
63803
1969
结果我被自己看到的震惊了
01:05
The pictures that I was finding
24
65772
1674
那些我看到的图像
01:07
were years out of date,
25
67446
1901
已经过时好多年了
01:09
and because of that,
26
69347
1068
也因为如此
01:10
it had relatively little relevance
27
70415
1671
它跟我当时的工作
01:12
to the work that I was doing today.
28
72086
2460
几乎没什么关联
01:14
But I was intrigued.
29
74546
1461
不过我很着迷
01:16
I mean, satellite imagery is pretty amazing stuff.
30
76007
3236
大家都知道,卫星图像是非常了不起的东西
01:19
There are millions and millions of sensors
31
79243
2303
今天我们身边有成千上百万的
01:21
surrounding us today,
32
81546
1444
传感器
01:22
but there's still so much we don't know on a daily basis.
33
82990
2960
然而每天都还有那么多 我们不知道的东西
01:25
How much oil is stored in all of China?
34
85950
3753
中国藏有多少石油?
01:29
How much corn is being produced?
35
89703
2933
有多少玉米被生产出来?
01:32
How many ships are in all of our world's ports?
36
92636
4071
全世界的所有港口有多少船只?
01:36
Now, in theory, all of these questions
37
96707
2559
理论上,所有的这些问题
01:39
could be answered by imagery,
38
99266
2244
都可以用图像来回答
01:41
but not if it's old.
39
101510
1727
不过如果是老的图像就不行了
01:43
And if this data was so valuable,
40
103237
2458
并且如果这些数据真的那么重要
01:45
then how come I couldn't get my hands
41
105695
1932
那为什么我不可以获得
01:47
on more recent pictures?
42
107627
2550
更新的图像呢?
01:50
So the story begins over 50 years ago
43
110177
3135
故事始于50多年前
01:53
with the launch of the first generation
44
113312
1871
美国政府发射的
01:55
of U.S. government photo reconnaissance satellites.
45
115183
3600
第一代照片侦查卫星
01:58
And today, there's a handful
46
118783
1696
如今,还有少数
02:00
of the great, great grandchildren
47
120479
2195
这些早期冷战时期的机器的
02:02
of these early Cold War machines
48
122674
1665
曾子孙们
02:04
which are now operated by private companies
49
124339
2009
它们现在多由私营公司运作
02:06
and from which the vast majority of satellite imagery
50
126348
2709
也是我们今天日常看到的
02:09
that you and I see on a daily basis comes.
51
129057
2825
绝大部分的卫星图像的来源
02:11
During this period, launching things into space,
52
131882
2914
在此期间,把东西发射到太空
02:14
just the rocket to get the satellite up there,
53
134796
2677
仅仅是用火箭将一个卫星送上去
02:17
has cost hundreds of millions of dollars each,
54
137473
4601
一次就要花掉数亿美元,十亿啊
02:22
and that's created tremendous pressure
55
142074
1752
这种强大的压力下 不能过于频繁地
02:23
to launch things infrequently
56
143826
2289
发射东西
02:26
and to make sure that when you do,
57
146115
1344
而当你发射东西的时候
02:27
you cram as much functionality in there as possible.
58
147459
3639
就要保证塞进去的功能越多越好
02:31
All of this has only made satellites
59
151098
1611
所有的这一切都导致了卫星
02:32
bigger and bigger and bigger
60
152709
2417
变得越来越大
02:35
and more expensive,
61
155126
1598
也越来越贵
02:36
now nearly a billion, with a b, dollars per copy.
62
156724
5037
如今每个要花掉近10亿美元
02:41
Because they are so expensive,
63
161761
1699
因为它们如此昂贵
02:43
there aren't very many of them.
64
163460
1467
所以数量就不是很多
02:44
Because there aren't very many of them,
65
164927
1345
因为数量不是很多
02:46
the pictures that we see on a daily basis
66
166272
2456
我们日常看到的图像
02:48
tend to be old.
67
168728
1504
就偏陈旧
02:50
I think a lot of people actually understand this anecdotally,
68
170232
3282
我想很多人可以理解这个故事
02:53
but in order to visualize just how sparsely
69
173514
2539
不过为了能让大家更好地想象
02:56
our planet is collected,
70
176053
1884
这些卫星图像有多么稀疏
02:57
some friends and I put together a dataset
71
177937
2066
我和我的一些朋友制作了一个数据库
03:00
of the 30 million pictures that have been gathered
72
180003
2587
这个数据库里有从2000年到2010年
03:02
by these satellites between 2000 and 2010.
73
182590
3484
这些卫星收集的3000万张图像
03:06
As you can see in blue, huge areas of our world
74
186074
2755
你可以看到蓝色的区域 全球有大面积的地方
03:08
are barely seen, less than once a year,
75
188829
2790
几乎看不到 图像少于一年一次
03:11
and even the areas that are seen most frequently,
76
191619
2272
即使是那些最经常看到的区域
03:13
those in red, are seen at best once a quarter.
77
193891
3748
红色的区域 也只是最多每个季度看到一次
03:17
Now as aerospace engineering grad students,
78
197639
2916
这时,作为航空航天工程专业的研究生
03:20
this chart cried out to us as a challenge.
79
200555
3428
这个图给我们发起了一个挑战
03:23
Why do these things have to be so expensive?
80
203983
3455
为什么这些东西要如此昂贵呢?
03:27
Does a single satellite really have to cost
81
207438
2735
单单一个卫星真的必须花费
03:30
the equivalent of three 747 jumbo jets?
82
210173
4763
等同于3架747大型喷气式客机的费用么?
03:34
Wasn't there a way to build a smaller,
83
214936
2163
难道没有办法造一个更小的
03:37
simpler, new satellite design that could enable
84
217099
3172
更简单的、更新的卫星
03:40
more timely imaging?
85
220271
2018
来提供即时的图像么?
03:42
I realize that it does sound a little bit crazy
86
222289
3451
我知道这听上去有些疯狂
03:45
that we were going to go out and just
87
225740
1511
我们准备自己动手
03:47
begin designing satellites,
88
227251
1847
开始设计卫星
03:49
but fortunately we had help.
89
229098
1963
不过幸运的是,我们有帮手
03:51
In the late 1990s, a couple of professors
90
231061
2935
在90年代后期,几个教授
03:53
proposed a concept for radically reducing the price
91
233996
3481
提出了一个概念,可以大幅降低
03:57
of putting things in space.
92
237477
1772
将东西发射到太空的成本
03:59
This was hitchhiking small satellites
93
239249
2111
即在发射大卫星的时候
04:01
alongside much larger satellites.
94
241360
3016
顺带稍上小卫星
04:04
This dropped the cost of putting objects up there
95
244376
2909
这让送东西上太空的成本
04:07
by over a factor of 100,
96
247285
1998
降低了100倍
04:09
and suddenly we could afford to experiment,
97
249283
3086
于是突然之间,我们可以做更多的实验
04:12
to take a little bit of risk,
98
252369
1468
冒更多的风险
04:13
and to realize a lot of innovation.
99
253837
2266
追求更多的创新
04:16
And a new generation of engineers and scientists,
100
256103
3089
新一代的工程师和科学家们
04:19
mostly out of universities,
101
259192
1599
他们大部分来自大学
04:20
began launching these very small,
102
260791
2343
开始打造这些很小的
04:23
breadbox-sized satellites called CubeSats.
103
263134
2591
面包盒大小的“立方体卫星”
04:25
And these were built with electronics obtained
104
265725
2403
这些卫星是用从 RadioShack公司(一家美国电子产品连锁店)
04:28
from RadioShack instead of Lockheed Martin.
105
268128
3898
而不是洛克希德·马丁公司(一家美国航空航天制造商) 买的电子部件打造的
04:32
Now it was using the lessons learned from these early missions
106
272026
2840
利用从这些早期发射中学到的经验
04:34
that my friends and I began a series of sketches
107
274866
2695
我和我的朋友开始设计
04:37
of our own satellite design.
108
277561
1689
我们自己的卫星
04:39
And I can't remember a specific day
109
279250
2784
我不记得具体是哪一天
04:42
where we made a conscious decision
110
282034
1860
我们突然做了这样一个有意识的决定
04:43
that we were actually going to go out and build these things,
111
283894
2635
就是我们真的要打造这些东西
04:46
but once we got that idea in our minds
112
286529
2430
不过一旦我们有了这些想法——
04:48
of the world as a dataset,
113
288959
2161
这个世界是一个数据库
04:51
of being able to capture millions of data points
114
291120
2490
每天我们都能捕捉数百万个数据点
04:53
on a daily basis describing the global economy,
115
293610
3053
来描绘全球经济
04:56
of being able to unearth billions of connections
116
296663
2520
我们能够发掘它们之间
04:59
between them that had never before been found,
117
299183
3165
从来没被发现过的、无数的关联——
05:02
it just seemed boring
118
302348
1648
其它一切的工作
05:03
to go work on anything else.
119
303996
2907
似乎都显得无趣
05:06
And so we moved into a cramped,
120
306903
2973
于是我们搬到了帕洛阿尔托 (美国加州的城市,科技创业公司的天堂)
05:09
windowless office in Palo Alto,
121
309876
2979
一个拥挤的、没有窗户的办公室
05:12
and began working to take our design
122
312855
1967
并开始将我们的设计
05:14
from the drawing board into the lab.
123
314822
2938
从图版转到实验室
05:17
The first major question we had to tackle
124
317760
2658
我们首先要对付的一大问题就是
05:20
was just how big to build this thing.
125
320418
2334
我们要造一个多大的东西
05:22
In space, size drives cost,
126
322752
3183
在太空,体积决定成本
05:25
and we had worked with these very small,
127
325935
2013
在学校,我们使用这些很小的、
05:27
breadbox-sized satellites in school,
128
327948
2094
面包盒大小的卫星
05:30
but as we began to better understand the laws of physics,
129
330042
2551
不过当我们开始更好地理解了物理定律
05:32
we found that the quality of pictures
130
332593
2122
我们发现这些卫星能够获取的
05:34
those satellites could take was very limited,
131
334715
3084
图像的质量非常局限
05:37
because the laws of physics dictate
132
337799
1704
因为物理定律决定了
05:39
that the best picture you can take through a telescope
133
339503
3106
通过望远镜拍的最好的照片
05:42
is a function of the diameter of that telescope,
134
342609
2239
是这个望远镜直径的函数
05:44
and these satellites had a very small,
135
344848
1662
而这些卫星只有很小的
05:46
very constrained volume.
136
346510
1706
非常有限的空间
05:48
And we found that the best picture we would
137
348216
1795
我们发现我们能够捕捉到的
05:50
have been able to get looked something like this.
138
350011
2485
最好的照片,看上去是这样的
05:52
Although this was the low-cost option,
139
352496
2245
虽然这是个低成本的选择
05:54
quite frankly it was just too blurry
140
354741
1673
但老实说,这实在太模糊了
05:56
to see the things that make satellite imagery valuable.
141
356414
3569
这样的卫星图像也没有价值
05:59
So about three or four weeks later,
142
359983
2146
大约三四个星期以后
06:02
we met a group of engineers randomly
143
362129
2817
我们巧遇到了一群工程师
06:04
who had worked on the first
144
364946
1641
他们曾经参与了第一架
06:06
private imaging satellite ever developed,
145
366587
2441
商业成像卫星的开发工作
06:09
and they told us that back in the 1970s,
146
369028
2118
他们告诉我们,过去在1970年代
06:11
the U.S. government had found a powerful
147
371146
2182
美国政府找到了一个强大的
06:13
optimal tradeoff --
148
373328
1643
最佳的权衡——
06:14
that in taking pictures at right about one meter resolution,
149
374971
3088
就是以刚好一米的分辨率进行拍照
06:18
being able to see objects one meter in size,
150
378059
2884
能够看到物体一米大小
06:20
they had found that they could not just get very high-quality images,
151
380943
2691
他们发现,他们不仅可以获得高质量的图像
06:23
but get a lot of them at an affordable price.
152
383634
2893
而且可以用低廉的价格获取很多
06:26
From our own computer simulations,
153
386527
1976
从我们自己的计算机模拟结果中
06:28
we quickly found that one meter really was
154
388503
2005
我们很快发现,一米确实是
06:30
the minimum viable product
155
390508
1764
最小的、可行的,能够看到
06:32
to be able to see the drivers of our global economy,
156
392272
3079
全球经济的驱动因素的分辨率
06:35
for the first time, being able to count
157
395351
1561
第一次,我们可以数
06:36
the ships and cars and shipping containers and trucks
158
396912
2721
那些在地球上每天在移动的
06:39
that move around our world on a daily basis,
159
399633
2824
船只、车辆、集装箱和卡车的数量
06:42
while conveniently still not being able to see individuals.
160
402457
3804
虽然还看不到清楚人物
06:46
We had found our compromise.
161
406261
1621
我们找到了我们的最佳权衡点
06:47
We would have to build something larger
162
407882
1604
我们需要造一个
06:49
than the original breadbox,
163
409486
1563
比原有的面包盒更大的东西
06:51
now more like a mini-fridge,
164
411049
1651
更像个迷你冰箱
06:52
but we still wouldn't have to build a pickup truck.
165
412700
2761
但是我们却不需要造一个货车
06:55
So now we had our constraint.
166
415461
2624
现在我们有了限制
06:58
The laws of physics dictated
167
418085
1757
物理定律决定了
06:59
the absolute minimum-sized telescope that we could build.
168
419842
3324
我们能够打造的最小的望远镜
07:03
What came next was making the rest of the satellite
169
423166
3054
接下来要做的,就是把这个卫星的其它部分
07:06
as small and as simple as possible,
170
426220
1741
造得尽可能地小而简单
07:07
basically a flying telescope with four walls
171
427961
2652
基本上就是一个会飞的望远镜,它有四面墙
07:10
and a set of electronics smaller than a phone book
172
430613
2941
和一些体积比通讯录还小的
07:13
that used less power than a 100 watt lightbulb.
173
433554
3438
用的电比100瓦灯泡还少的电子部件
07:16
The big challenge became actually taking
174
436992
1900
真正最大的挑战是
07:18
the pictures through that telescope.
175
438892
2198
通过那个望远镜拍照
07:21
Traditional imaging satellites use a line scanner,
176
441090
3044
传统的成像卫星采用行扫描仪
07:24
similar to a Xerox machine,
177
444134
1761
类似于复印机
07:25
and as they traverse the Earth, they take pictures,
178
445895
2400
随着它们穿越地球,它们进行拍照
07:28
scanning row by row by row
179
448295
2196
逐行地进行扫描
07:30
to build the complete image.
180
450491
1675
从而建立完整的图像
07:32
Now people use these because they get a lot of light,
181
452166
3004
如今人们用这个,因为它们可以获得很多光线
07:35
which means less of the noise you see
182
455170
1831
也就意味着少一些
07:37
in a low-cost cell phone image.
183
457016
2736
你在便宜的手机照片上会看到的噪点
07:39
The problem with them is they require
184
459752
2361
这个方法的问题是,它们需要
07:42
very sophisticated pointing.
185
462113
2240
非常复杂的定位
07:44
You have to stay focused on a 50-centimeter target
186
464353
2546
你需要从600英里以外
07:46
from over 600 miles away
187
466899
1752
聚焦一个50厘米的对象
07:48
while moving at more than seven kilometers a second,
188
468651
2232
同时以每秒7公里的速度移动
07:50
which requires an awesome degree of complexity.
189
470883
2749
这些都需要非常高的准确度
07:53
So instead, we turned to a new generation of video sensors,
190
473632
3437
因此,我们采用了一种新的视频传感器
07:57
originally created for use in night vision goggles.
191
477069
3091
它最早是用于夜视镜的
08:00
Instead of taking a single, high quality image,
192
480160
2899
我们不拍单一的高质量图像
08:03
we could take a videostream
193
483059
1387
而是拍一段
08:04
of individually noisier frames,
194
484446
2923
由很多帧噪点较高的图像组成的视频
08:07
but then we could recombine
195
487369
1744
然后我们在地面上
08:09
all of those frames together
196
489113
1466
运用复杂的像素处理技术
08:10
into very high-quality images
197
490579
1996
把这些图像重新组合在一起
08:12
using sophisticated pixel processing techniques
198
492575
2593
把这些图像重新组合在一起
08:15
here on the ground,
199
495168
1521
组成高质量的图像
08:16
at a cost of one one hundredth a traditional system.
200
496689
3137
而花费只是传统办法的百分之一
08:19
And we applied this technique
201
499826
1153
我们还把这个技术
08:20
to many of the other systems on the satellite as well,
202
500979
2974
运用到卫星的很多其它系统上
08:23
and day by day, our design evolved
203
503953
2840
一天天过去,我们的设计已经从
08:26
from CAD to prototypes
204
506793
3489
CAD(计算机辅助设计)变成了样机
08:30
to production units.
205
510282
2888
从样机变成了实际产品
08:33
A few short weeks ago,
206
513170
1657
短短几个星期前
08:34
we packed up SkySat 1,
207
514827
1851
我们装好了Skysat1
08:36
put our signatures on it,
208
516678
1773
我们在上面签了名
08:38
and waved goodbye for the last time on Earth.
209
518451
2223
最后一次在地球上跟它挥手告别
08:40
Today, it's sitting in its final launch configuration
210
520674
3481
今天,它正坐在它的最终发射装置里
08:44
ready to blast off in a few short weeks.
211
524155
3061
准备在几个星期后升空
08:47
And soon, we'll turn our attention to launching
212
527216
2342
很快,我们会转向发射
08:49
a constellation of 24 or more of these satellites
213
529558
3031
一系列的卫星,24个或者更多
08:52
and beginning to build the scalable analytics
214
532589
2550
并开始建立可扩展的分析
08:55
that will allow us to unearth the insights
215
535139
2339
使得我们能够从我们收集的
08:57
in the petabytes of data we will collect.
216
537478
3189
拍字节(10的15次方)的数据中发掘信息
09:00
So why do all of this? Why build these satellites?
217
540667
3973
那么为什么要做这些呢?为什么要造这些卫星?
09:04
Well, it turns out imaging satellites
218
544640
2611
因为成像卫星
09:07
have a unique ability to provide global transparency,
219
547251
3710
有提供全球透明度的独特能力
09:10
and providing that transparency on a timely basis
220
550961
2626
而即时提供
09:13
is simply an idea whose time has come.
221
553587
3411
全球透明度的时代已经到来
09:16
We see ourselves as pioneers of a new frontier,
222
556998
3523
我们把自己看作是新领域的先驱者
09:20
and beyond economic data,
223
560521
1764
在经济数据以外
09:22
unlocking the human story, moment by moment.
224
562285
3601
慢慢解开人类之谜
09:25
For a data scientist
225
565886
1563
对一个碰巧
09:27
that just happened to go to space camp as a kid,
226
567449
2750
孩提时代去了太空营的数据科学家而言
09:30
it just doesn't get much better than that.
227
570199
2715
没有什么比这更好的了
09:32
Thank you.
228
572914
2215
谢谢大家!
09:35
(Applause)
229
575129
4134
(掌声)
关于本网站

这个网站将向你介绍对学习英语有用的YouTube视频。你将看到来自世界各地的一流教师教授的英语课程。双击每个视频页面上显示的英文字幕,即可从那里播放视频。字幕会随着视频的播放而同步滚动。如果你有任何意见或要求,请使用此联系表与我们联系。

https://forms.gle/WvT1wiN1qDtmnspy7