AMRDEC Presentation V0.

Report
ECE 497
Capstone Design
Will Barrett Asato Tashiro Adam Anderson
Purpose of the System
• The purpose is to create a scanning system to
determine the location and size of metal
fragments in a medium density fiber panel.
Background Information
• The Weapons Integration & Development
Directorate of the US Army Aviation and Missile
Research, Development, & Engineering Center
performs a variety of munitions and warhead test
programs
– Known as the “AMRDEC”
– Patrick Taylor is our sponsor
• They use bundles of fiber panels to capture
shrapnel from the explosions
• Each bundle is then searched BY HAND to recover
fragments, tabulate the X/Y position, and mass
for each panel
– The panel location is used to determine the depth of
the fragment
Background Information (cont’d)
• The process can take up to 100 man hours PER PANEL
• Each experiment could require up to 96 panels, or 9600
man-hours per experiment
• Removal is done outside on the test range
– Personnel have to be in protective clothing
– Fragments are bagged by weight
– Fragments have to be cleaned and decontaminated for safe
handling
• Fragments are typically steel, but some tests use titanium
or aluminum
• Data derived from the analysis is tabulated in a spreadsheet
format, with the fragment designation, count, bundle and
panel number, X/Y location, and size/weight
Fragment Set Weight Distribution
STEEL WARHEAD FRAGMENT SET
FOR FRAG BUNDLE SCANNER TESTING
Totals\Wt Range
Quantity
Total Wt. (g)
Avg. Wt. (g)
% of Total Captured Wt.
<.25g
235
34.185
0.145
2.89%
.25g - .75g .75g - 2g 2g - 5g
272
271
157
130.017 346.711 467.746
0.478
1.279
2.979
10.99% 29.29% 39.52%
5g - 7g
19
109.794
5.779
39.52%
7g - 10g
7
64.433
8.919
9.28%
10+g
3
32.659
10.886
2.76%
Typical Fragment Morphology
Project Objective
• Automate analysis procedures to the
maximum degree feasible
– Analysis of X/Y location goal is 15 minutes/panel
• Portability – system can be setup by 1 person
• Battery operated equipment is desirable
Measurement Parameters
• X/Y resolution of .5”
• Minimum fragment detection of .25g
Approach
• Demonstrate a proof-of-concept capability
that can be scaled up at a later time
• Use commercial-off-the-shelf (COTS)
equipment to produce a gray-scale image of
the fragments contained in a panel
• Process image to identify the centroid of each
fragment and X/Y location
– Output .csv file with the fragment location data
COTS Imaging Systems
• Ground Penetrating Radar (GPR)
• Industrial Radiography
What is Ground Penetrating Radar?
• GPR is a non-destructive imaging method that uses
radar pulses to image the subsurface
• The radar pulses used are in the microwave band of
the radio spectrum
How does GPR work?
• GPR is similar to a metal detector
• The GPR system sends out thousands of RF
frequency pulses into the ground
• The frequencies that are reflected back return to the
antennae
• Frequency analysis of the reflected RF energy allows
correlation of different material compositions in the
subsurface
GPR Schematic
Typical Applications
•
•
•
•
Pipe Locating
Archeology
IED Locating (Improvised Explosive Devices)
Quality control of reinforced concrete
Ground Penetrating Radar
GPR Scans
Describing Industrial Radiography
• Industrial Radiography is also a non-destructive
detection method which utilizes X-rays and gamma
rays to view materials
How Industrial Radiography Works
• Similar to medical radiography, Industrial radiography
uses an X-ray source to bombard a sample with highenergy radiation onto a film or a digital detector
• This creates a 2D image of the different materials in
the sample.
Radiography Schematic
Typical Applications
• Security
• Medical Imaging
• Non-Destructive Testing
– Castings
– Welds
Medical X-Ray Radiography
X-Ray Source
Imaging Surface
Industrial Radiography
• (Insert pic of Xerox)
Program Plan
• Contact vendors
– Image dummy panels using radiography and GPR
– Utilize 2D images for image processing study
• Image Processing Approach
– 2D grayscale image can be processed using
MATLAB to determine the centroid and area of
each fragment
– Tabulate fragments by area and X/Y location
Current Status
• Dummy panels have been fabricated
• They will be shipped/delivered to USRadar,
Hayes, ATS, and University Hospital this week
• Start analysis of our simulated fragment panel
this week
Simulated Fragment Panel
Simulated Fragment Panel
Black and White
Simulated Panel with Border Detection
Image Analysis Data
Pixcavator 5.1
Dark objects:
Light objects
61
0
OUTPUT
ID
Type Location X
Location Y
Size
Perim.
Round.
Gray
Contrast
Thick.
Length
ID
1
2
D
D
192
105
110
140
210
225
59
62
76
72
44
45
211
210
11
11
17
20
1
2
3
D
530
147
198
54
84
38
217
13
13
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
324
471
270
164
540
270
139
434
229
407
140
54
346
110
319
510
353
257
309
329
365
412
241
183
443
408
382
426
413
430
429
417
418
392
638
1025
123
168
115
364
236
477
3778
4338
816
1018
362
253
287
1287
322
110
151
43
53
42
88
61
87
377
414
201
164
77
62
74
165
77
66
56
81
74
80
58
77
78
33
32
25
47
76
80
65
59
67
28
23
63
52
50
40
43
31
17
16
39
27
34
36
40
21
35
227
232
192
203
205
215
212
224
238
239
216
228
221
219
215
234
220
16
17
10
10
10
10
13
19
22
23
8
15
15
15
10
20
12
38
57
10
16
10
33
17
23
165
183
91
67
22
15
26
61
26
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

similar documents