1
00:00:04,150 --> 00:00:01,990
this video is a response to a discussion

2
00:00:06,789 --> 00:00:04,160
between former fighter pilot chris lato

3
00:00:07,990 --> 00:00:06,799
and thermal camera technician dave fulch

4
00:00:10,310 --> 00:00:08,000
in that discussion

5
00:00:12,070 --> 00:00:10,320
dave claimed to debunk my rotating glare

6
00:00:14,230 --> 00:00:12,080
hypothesis for the gimbal video and

7
00:00:15,829 --> 00:00:14,240
chris claims to understand this

8
00:00:17,349 --> 00:00:15,839
i'll demonstrate that neither claim is

9
00:00:19,269 --> 00:00:17,359
true

10
00:00:21,029 --> 00:00:19,279
but first let's start by establishing

11
00:00:23,509 --> 00:00:21,039
some facts

12
00:00:25,509 --> 00:00:23,519
firstly this is an example of infrared

13
00:00:27,750 --> 00:00:25,519
glare that obscures an entire plane

14
00:00:29,589 --> 00:00:27,760
it's from the chilean navy ufo case that

15
00:00:34,150 --> 00:00:29,599
was solved in 2017.

16
00:00:36,470 --> 00:00:34,160
it's an airbus a340 iberia flight 6830

17
00:00:37,430 --> 00:00:36,480
in infrared we see only large glares

18
00:00:39,670 --> 00:00:37,440
around the engine

19
00:00:41,830 --> 00:00:39,680
and we can't see the plane so infrared

20
00:00:44,470 --> 00:00:41,840
glare exists

21
00:00:46,310 --> 00:00:44,480
secondly this is a narrow field of view

22
00:00:48,630 --> 00:00:46,320
a thousand millimeter focal length the

23
00:00:50,470 --> 00:00:48,640
plane is over 60 miles away so infrared

24
00:00:52,470 --> 00:00:50,480
layer that obscures an airplane exists

25
00:00:54,709 --> 00:00:52,480
in narrow field of view

26
00:00:55,830 --> 00:00:54,719
thirdly the shape of a glare is relative

27
00:00:57,750 --> 00:00:55,840
to the camera

28
00:00:59,830 --> 00:00:57,760
so if you rotate the camera the scene

29
00:01:01,990 --> 00:00:59,840
rotates but the glare will stay in the

30
00:01:03,430 --> 00:01:02,000
same orientation in the video

31
00:01:05,270 --> 00:01:03,440
you can demonstrate this with a phone

32
00:01:07,030 --> 00:01:05,280
pointing at a flashlight rotate the

33
00:01:09,670 --> 00:01:07,040
phone and in the video the background

34
00:01:12,390 --> 00:01:09,680
rotates but the glare does not

35
00:01:15,030 --> 00:01:12,400
fourthly a derotation mechanism will

36
00:01:16,230 --> 00:01:15,040
rotate the entire image glare and all

37
00:01:17,749 --> 00:01:16,240
and will have no effect on the

38
00:01:19,109 --> 00:01:17,759
orientation of the glare within the

39
00:01:21,749 --> 00:01:19,119
scene

40
00:01:23,590 --> 00:01:21,759
what does that mean well the aptly a

41
00:01:25,190 --> 00:01:23,600
camera system is mounted inside a pod

42
00:01:25,749 --> 00:01:25,200
that must rotate to keep tracking an

43
00:01:27,590 --> 00:01:25,759
object

44
00:01:29,670 --> 00:01:27,600
but the pilot does not want the image on

45
00:01:30,310 --> 00:01:29,680
screen rotating they want the horizon on

46
00:01:32,390 --> 00:01:30,320
their screen

47
00:01:34,469 --> 00:01:32,400
to match the horizon through the window

48
00:01:35,429 --> 00:01:34,479
so there's a d rotation mechanism built

49
00:01:37,510 --> 00:01:35,439
in to correct that

50
00:01:40,789 --> 00:01:37,520
it simply rotates the entire image back

51
00:01:43,830 --> 00:01:40,799
so the horizon is in the right place

52
00:01:45,670 --> 00:01:43,840
finally if you derotate an image in

53
00:01:47,030 --> 00:01:45,680
which the background has rotated but the

54
00:01:48,469 --> 00:01:47,040
glare has not rotated

55
00:01:50,550 --> 00:01:48,479
then it's going to look like the glare

56
00:01:52,069 --> 00:01:50,560
rotates but the background doesn't

57
00:01:53,670 --> 00:01:52,079
we can demonstrate that by using the

58
00:01:56,230 --> 00:01:53,680
same video from earlier

59
00:01:57,749 --> 00:01:56,240
the simplest way to de-rotate it is just

60
00:01:59,670 --> 00:01:57,759
to film it with another camera

61
00:02:01,749 --> 00:01:59,680
now the image remains stable but the

62
00:02:04,870 --> 00:02:01,759
glare seems to rotate

63
00:02:06,950 --> 00:02:04,880
to recap infrared glare exists it can

64
00:02:08,150 --> 00:02:06,960
cover a plane it exists even in narrow

65
00:02:10,229 --> 00:02:08,160
field of view

66
00:02:12,070 --> 00:02:10,239
glare shape is relative to the camera

67
00:02:13,110 --> 00:02:12,080
rotating the camera rotates the scene

68
00:02:14,710 --> 00:02:13,120
but not the glare

69
00:02:17,750 --> 00:02:14,720
so the angle of the glare relative to

70
00:02:20,869 --> 00:02:17,760
the scene changes derotating the image

71
00:02:22,550 --> 00:02:20,879
rotates everything the same amount

72
00:02:24,390 --> 00:02:22,560
derotating an image where the background

73
00:02:26,150 --> 00:02:24,400
has rotated but the glare has not will

74
00:02:30,150 --> 00:02:26,160
make it look like the background has not

75
00:02:33,030 --> 00:02:30,160
rotated but the glare has rotated

76
00:02:33,670 --> 00:02:33,040
okay now let's get into dave's debunking

77
00:02:35,270 --> 00:02:33,680
first of all

78
00:02:38,630 --> 00:02:35,280
dave agrees with me about the function

79
00:02:42,070 --> 00:02:38,640
of the derotation mechanism

80
00:02:44,790 --> 00:02:42,080
why do you need that i guess like

81
00:02:45,910 --> 00:02:44,800
let's say you have your gimbal when it's

82
00:02:47,589 --> 00:02:45,920
rotating

83
00:02:49,190 --> 00:02:47,599
you don't want the scene to rotate to

84
00:02:51,990 --> 00:02:49,200
the side so you have a d

85
00:02:53,990 --> 00:02:52,000
rotation counteract that it's going to

86
00:02:55,670 --> 00:02:54,000
rotate the whole scene

87
00:02:57,910 --> 00:02:55,680
okay so we're all in agreement the

88
00:02:58,390 --> 00:02:57,920
derotation mechanism rotates the entire

89
00:02:59,830 --> 00:02:58,400
scene

90
00:03:02,309 --> 00:02:59,840
and it's there to keep the horizon in

91
00:03:03,670 --> 00:03:02,319
the right place so let's see dave's

92
00:03:06,149 --> 00:03:03,680
objection

93
00:03:07,110 --> 00:03:06,159
the big thing is refraction versus

94
00:03:09,830 --> 00:03:07,120
reflection

95
00:03:10,309 --> 00:03:09,840
because he likes talking about how the

96
00:03:15,030 --> 00:03:10,319
glare

97
00:03:16,309 --> 00:03:15,040
about it's um

98
00:03:18,070 --> 00:03:16,319
it's something to do where there's

99
00:03:20,790 --> 00:03:18,080
already an existing glare

100
00:03:21,430 --> 00:03:20,800
and then he theorizes that the gimbal

101
00:03:23,270 --> 00:03:21,440
moves

102
00:03:25,509 --> 00:03:23,280
and then like a derotation device

103
00:03:28,149 --> 00:03:25,519
counteracts the rotation

104
00:03:28,630 --> 00:03:28,159
of the gimbal moving but what i'm saying

105
00:03:34,710 --> 00:03:28,640
is

106
00:03:35,910 --> 00:03:34,720
it's going to rotate the whole scene

107
00:03:38,869 --> 00:03:35,920
opposed to

108
00:03:39,910 --> 00:03:38,879
refracting if you have a lens like this

109
00:03:43,190 --> 00:03:39,920
this is like a

110
00:03:44,869 --> 00:03:43,200
silicone then you might have some

111
00:03:47,350 --> 00:03:44,879
refraction with the glare

112
00:03:49,670 --> 00:03:47,360
you have the lens moving but the d

113
00:03:51,990 --> 00:03:49,680
rotation device that he touches on

114
00:03:53,110 --> 00:03:52,000
it's it's not that at all it's it's a

115
00:03:55,830 --> 00:03:53,120
couple mirrors

116
00:03:56,789 --> 00:03:55,840
like this that are going to rotate the

117
00:03:59,350 --> 00:03:56,799
whole scene

118
00:04:00,390 --> 00:03:59,360
so the object can't independently rotate

119
00:04:02,190 --> 00:04:00,400
on its own

120
00:04:03,910 --> 00:04:02,200
magically like this glare that he's

121
00:04:06,710 --> 00:04:03,920
hypothesizing with

122
00:04:08,550 --> 00:04:06,720
follow that he says the derotation

123
00:04:09,830 --> 00:04:08,560
device rotates the whole scene which is

124
00:04:11,750 --> 00:04:09,840
what i say

125
00:04:13,270 --> 00:04:11,760
then he says as opposed to refracting

126
00:04:14,949 --> 00:04:13,280
which doesn't really make any sense but

127
00:04:16,550 --> 00:04:14,959
it's also irrelevant as it just agreed

128
00:04:17,909 --> 00:04:16,560
with me that the de-rotation mechanism

129
00:04:20,469 --> 00:04:17,919
rotates the entire scene

130
00:04:22,310 --> 00:04:20,479
including the glare by the same amount

131
00:04:24,870 --> 00:04:22,320
remember what we established earlier

132
00:04:27,350 --> 00:04:24,880
glare shape is relative to the camera if

133
00:04:29,909 --> 00:04:27,360
the camera rotates the scene will rotate

134
00:04:31,030 --> 00:04:29,919
but the glare will not derotating the

135
00:04:32,710 --> 00:04:31,040
entire image

136
00:04:36,070 --> 00:04:32,720
makes it looks like the scene is not

137
00:04:39,110 --> 00:04:36,080
rotating but the glare is rotating

138
00:04:41,430 --> 00:04:39,120
how does chris respond to this

139
00:04:43,510 --> 00:04:41,440
so the way i understand it is that it's

140
00:04:44,550 --> 00:04:43,520
coming in and so there's a glare somehow

141
00:04:48,550 --> 00:04:44,560
on the lens

142
00:04:50,790 --> 00:04:48,560
that but it's not it it rotates

143
00:04:51,909 --> 00:04:50,800
right and so the actual scent the image

144
00:04:53,670 --> 00:04:51,919
is is the same

145
00:04:56,070 --> 00:04:53,680
right it's seeing the image but since

146
00:04:58,150 --> 00:04:56,080
the glass is rotating on top of it

147
00:04:59,909 --> 00:04:58,160
that changes just the glare but not

148
00:05:02,070 --> 00:04:59,919
really all the other image which is seen

149
00:05:03,990 --> 00:05:02,080
normally that's the way i understand it

150
00:05:05,510 --> 00:05:04,000
well kind of glare isn't something

151
00:05:07,510 --> 00:05:05,520
that's on the lens

152
00:05:09,670 --> 00:05:07,520
glare is a scattering of light or

153
00:05:10,469 --> 00:05:09,680
radiation around a very bright radiation

154
00:05:12,469 --> 00:05:10,479
source

155
00:05:14,390 --> 00:05:12,479
a lens window or mirror has

156
00:05:15,990 --> 00:05:14,400
imperfections so a small portion of the

157
00:05:17,110 --> 00:05:16,000
light gets scattered

158
00:05:18,870 --> 00:05:17,120
the amount and direction of this

159
00:05:20,710 --> 00:05:18,880
scattering is defined by the glare

160
00:05:22,310 --> 00:05:20,720
spread function which is also called the

161
00:05:23,670 --> 00:05:22,320
point spread function

162
00:05:25,670 --> 00:05:23,680
but i think what he's trying to convey

163
00:05:27,430 --> 00:05:25,680
here is that the glare has already

164
00:05:29,830 --> 00:05:27,440
rotated relative to the horizon because

165
00:05:32,550 --> 00:05:29,840
the camera has rotated

166
00:05:33,350 --> 00:05:32,560
dave's response to this is interesting

167
00:05:35,350 --> 00:05:33,360
okay

168
00:05:36,629 --> 00:05:35,360
we'll touch on that um there's going to

169
00:05:38,710 --> 00:05:36,639
be two different

170
00:05:39,670 --> 00:05:38,720
sensors there's going to be a daylight

171
00:05:41,749 --> 00:05:39,680
camera sensor

172
00:05:43,510 --> 00:05:41,759
which is going to utilize the sapphire

173
00:05:44,710 --> 00:05:43,520
glass then there's going to be the

174
00:05:46,390 --> 00:05:44,720
infrared sensor

175
00:05:48,550 --> 00:05:46,400
that's it's going to have this kind of

176
00:05:51,749 --> 00:05:48,560
acid germanium or silicone or

177
00:05:53,110 --> 00:05:51,759
some other type of semiconductor

178
00:05:55,830 --> 00:05:53,120
metalloid

179
00:05:58,309 --> 00:05:55,840
so what what they're doing there is heat

180
00:06:01,189 --> 00:05:58,319
doesn't travel through just glass

181
00:06:02,629 --> 00:06:01,199
sapphire glass it actually blocks it so

182
00:06:04,230 --> 00:06:02,639
if you're trying to run from the cops

183
00:06:05,350 --> 00:06:04,240
one of your best places to hide is

184
00:06:07,029 --> 00:06:05,360
probably a greenhouse

185
00:06:09,270 --> 00:06:07,039
just nothing but glass sounds legit

186
00:06:11,670 --> 00:06:09,280
right as dave is the ir expert right

187
00:06:13,430 --> 00:06:11,680
wrong the lockheed martin sniper

188
00:06:14,390 --> 00:06:13,440
targeting pot that chris is probably

189
00:06:16,790 --> 00:06:14,400
familiar with

190
00:06:18,550 --> 00:06:16,800
actually uses a sapphire glass window

191
00:06:20,309 --> 00:06:18,560
specifically because sapphire glass is

192
00:06:23,029 --> 00:06:20,319
transparent to both visible light

193
00:06:24,230 --> 00:06:23,039
and medium wave infrared the lantern and

194
00:06:26,150 --> 00:06:24,240
at-flare parts use

195
00:06:27,830 --> 00:06:26,160
zinc sulfide which is similar to

196
00:06:29,189 --> 00:06:27,840
germanium in that it's transparent to

197
00:06:31,350 --> 00:06:29,199
the long wave infrared

198
00:06:33,670 --> 00:06:31,360
however it's also transparent to visible

199
00:06:35,189 --> 00:06:33,680
light germanium is opaque

200
00:06:37,430 --> 00:06:35,199
so that dave's misunderstanding

201
00:06:38,950 --> 00:06:37,440
something fairly significant

202
00:06:41,110 --> 00:06:38,960
so dave's objection doesn't really make

203
00:06:42,790 --> 00:06:41,120
any sense until his credit chris later

204
00:06:45,189 --> 00:06:42,800
admits he didn't really understand what

205
00:06:47,029 --> 00:06:45,199
dave is saying and asked him to clarify

206
00:06:48,309 --> 00:06:47,039
i understand that argument actually can

207
00:06:50,950 --> 00:06:48,319
you go through that again

208
00:06:53,350 --> 00:06:50,960
please so the d rotation device is

209
00:06:54,950 --> 00:06:53,360
something similar to this you can see

210
00:06:58,070 --> 00:06:54,960
when i turn it it rotates the whole

211
00:07:00,309 --> 00:06:58,080
scene it ro it rotates

212
00:07:03,510 --> 00:07:00,319
opposite the sensor so that the sensor

213
00:07:05,270 --> 00:07:03,520
always sees the correct up and down

214
00:07:06,629 --> 00:07:05,280
is that what you're saying sure and

215
00:07:08,870 --> 00:07:06,639
that's what his big

216
00:07:10,710 --> 00:07:08,880
argument was at first until i showed him

217
00:07:14,950 --> 00:07:10,720
on with it using his own

218
00:07:15,510 --> 00:07:14,960
um example um the atlar pdf i think he

219
00:07:18,309 --> 00:07:15,520
found

220
00:07:20,070 --> 00:07:18,319
online somewhere that should like um

221
00:07:21,029 --> 00:07:20,080
what the internals were like internal

222
00:07:22,790 --> 00:07:21,039
schematics

223
00:07:24,150 --> 00:07:22,800
and i said well that derotation device

224
00:07:26,230 --> 00:07:24,160
it explains that it's

225
00:07:27,430 --> 00:07:26,240
it's a mirror it's an infrared mirror

226
00:07:28,070 --> 00:07:27,440
which you're not going to have on

227
00:07:31,189 --> 00:07:28,080
anything

228
00:07:31,670 --> 00:07:31,199
like this these are all internal optics

229
00:07:34,790 --> 00:07:31,680
of

230
00:07:36,550 --> 00:07:34,800
app fleer or fliers that um

231
00:07:38,230 --> 00:07:36,560
yeah you're not gonna have any exposure

232
00:07:40,870 --> 00:07:38,240
to them unless you're somebody

233
00:07:42,469 --> 00:07:40,880
like myself or john that makes any sense

234
00:07:44,070 --> 00:07:42,479
notice he doesn't really make any points

235
00:07:45,909 --> 00:07:44,080
here and then ends it with an argument

236
00:07:47,430 --> 00:07:45,919
to authority the john he refers to is

237
00:07:48,950 --> 00:07:47,440
john earhart who had a similar

238
00:07:50,469 --> 00:07:48,960
misunderstanding that i covered in

239
00:07:52,390 --> 00:07:50,479
another video

240
00:07:54,070 --> 00:07:52,400
chris recognizes both this appeal to

241
00:07:55,510 --> 00:07:54,080
authority and the lack of a point and

242
00:07:57,749 --> 00:07:55,520
presses him some more

243
00:07:59,350 --> 00:07:57,759
basically you're saying that well first

244
00:08:02,390 --> 00:07:59,360
you actually work on these things so

245
00:08:02,869 --> 00:08:02,400
you you can be considered to know where

246
00:08:04,710 --> 00:08:02,879
you

247
00:08:06,790 --> 00:08:04,720
at least the internals but now that

248
00:08:07,749 --> 00:08:06,800
derotation device so the fact that that

249
00:08:10,790 --> 00:08:07,759
that

250
00:08:13,510 --> 00:08:10,800
argument

251
00:08:13,909 --> 00:08:13,520
and dave fails to provide any answer

252
00:08:17,110 --> 00:08:13,919
because

253
00:08:17,990 --> 00:08:17,120
he was thinking that the d rotation

254
00:08:20,230 --> 00:08:18,000
device would have

255
00:08:21,350 --> 00:08:20,240
was something like this that was a lens

256
00:08:22,950 --> 00:08:21,360
that would turn

257
00:08:25,029 --> 00:08:22,960
and so if you already have an existing

258
00:08:27,589 --> 00:08:25,039
glare then this lens

259
00:08:28,629 --> 00:08:27,599
could rotate the glare because of the

260
00:08:31,270 --> 00:08:28,639
refraction

261
00:08:31,670 --> 00:08:31,280
opposed to the reflection of a mirror

262
00:08:34,230 --> 00:08:31,680
chris

263
00:08:35,110 --> 00:08:34,240
unfortunately pretty much gives up here

264
00:08:38,149 --> 00:08:35,120
i get it

265
00:08:39,350 --> 00:08:38,159
okay i understand no you don't

266
00:08:41,029 --> 00:08:39,360
understand

267
00:08:42,870 --> 00:08:41,039
so once you see that dave really doesn't

268
00:08:44,230 --> 00:08:42,880
have a point with a d rotation or a

269
00:08:46,070 --> 00:08:44,240
point with the refraction versus

270
00:08:46,870 --> 00:08:46,080
reflection or a point with glass versus

271
00:08:48,389 --> 00:08:46,880
germanium

272
00:08:49,910 --> 00:08:48,399
then all that's left is this argument

273
00:08:51,350 --> 00:08:49,920
that you don't get glare in ir

274
00:08:53,269 --> 00:08:51,360
like this but we've already seen with

275
00:08:55,030 --> 00:08:53,279
the chilean case that you do

276
00:08:56,949 --> 00:08:55,040
this unfortunately has been an argument

277
00:08:58,949 --> 00:08:56,959
that's been going around for years

278
00:09:00,630 --> 00:08:58,959
dave is ostensibly an expert even though

279
00:09:02,230 --> 00:09:00,640
he's never worked on targeting pods

280
00:09:03,750 --> 00:09:02,240
and he didn't know that sapphire glasses

281
00:09:05,509 --> 00:09:03,760
are transparent

282
00:09:07,590 --> 00:09:05,519
but he doesn't really understand my

283
00:09:09,269 --> 00:09:07,600
arguments and since his debunking

284
00:09:11,190 --> 00:09:09,279
doesn't really make any sense then

285
00:09:13,030 --> 00:09:11,200
chris's claim that he understands is