1 automatic target recognition (atr) from the book: machine vision authors: w.e.snyder & h.qi...

37
1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

Upload: aldous-riley

Post on 21-Jan-2016

228 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

1

Automatic Target Recognition (ATR)

From the book: Machine VisionAuthors: W.E.Snyder & H.Qi

Cambridge University Press, 2004

Page 2: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

2

Automatic Target Recognition (ATR)

Page 3: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

3

Automatic Target Recognition (ATR)

Page 4: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

4

Automatic Target Recognition (ATR)

Page 5: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

5

Performance Evaluation of Automatic Target Recognition (ATR)

Page 6: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

6

Performance Evaluation of Automatic Target Recognition (ATR)

Page 7: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

7

Performance Evaluation of Automatic Target Recognition (ATR)

Page 8: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

8

Performance Evaluation of Automatic Target Recognition (ATR)

Page 9: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

9

Performance Evaluation of Automatic Target Recognition (ATR)

حالت های مختلف در ارتباط با تشخیص وجود هدف در صحنه

Page 10: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

10

Performance Evaluation of Automatic Target Recognition (ATR)

حالت های مختلف در ارتباط با تشخیص وجود هدف در صحنه

سیستم تشخیص ما اشتباه Falseتوجه کنید که در هر دو حالت کرده است.

Page 11: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

11

Performance Evaluation of Automatic Target Recognition (ATR)

Page 12: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

12

Performance Evaluation of Automatic Target Recognition (ATR)

Page 13: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

13

Performance Evaluation of Automatic Target Recognition (ATR)

Page 14: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

14

Performance Evaluation of Automatic Target Recognition (ATR)

الزم sensitivity, Specificityتوجه کنید که برای محاسبه مقادیر است که سیستم ، تعداد دفعات قابل قبولی مورد آزمایش

قرار گیرد تا مقادیر عددی که نشان دهنده تعداد دفعات وقوع هر حالت است بدست آید.

تعداد دفعات آزمایش بستگی به امکانات جمع آوری داده ها در سیستم مورد بر رسی دارد.

Page 15: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

15

بايد داراي چه توانائي ATRالگوريتم هاي هائي باشند

Page 16: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

16

بايد داراي چه توانائي ATRالگوريتم هاي هائي باشند

اغلب ( Occluded targets)- مسئله هم پوشانی 2اتفاق

می افتد.

Page 17: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

17

بايد داراي چه توانائي ATRالگوريتم هاي هائي باشند

Page 18: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

18

بايد داراي چه توانائي ATRالگوريتم هاي هائي باشند

Page 19: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

19

بايد داراي چه توانائي ATRالگوريتم هاي هائي باشند

Page 20: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

20

بايد داراي چه توانائي ATRالگوريتم هاي هائي باشند

Page 21: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

21

Target signature variability and false alarm rate

امضا ء(signature) يك هدف ميتواند يك امضا هندسي )شكل هدف و يا يك ويژه گي از شكل آن( و يا يك

باشد ، که بسته به جنس (Spectral)امضاء نوري سطح هدف میتوان بازتاب های متفاوتی ایجاد نماید.

، معموال تغييرات در بازتاب نور از سطح جسمتغييرات نور محيط ، هم پوشاني هدف از مشكالت

عمده در اين زمينه هستند.

Page 22: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

22

Target signature variability and false alarm rate

بطور كلي سيگنال ثبت شده در يك تصوير حاصل جمع نورصاطح شده از جسم و نور بازتاب يافته از آن است:

البته مقدار تابش ممكن است از مقدار بازتاب بسيار كمتر باشد )در شرايطنور قابل رويت( ، ويا اينكه بسيار قابل مالحظه باشد ) در شرايط نور

ماوراي بنفش با طول موج بلند(.

در اين حالتemissivity.بازتاب( از رابطه زير بدست مي آيد(

),(),(),( yxyxyxf

radiationtotal

radiationemitted

yxf

yxyx

),(

),(),(

Page 23: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

23

Target signature variability and false alarm rate

.نور صاطح شده با درجه حرارت مرتبط است

Aمسا حت سطح كه به متر مربع اندازه گرفته ميشود ثابت Stefan constant = 5.67x 10-8 W m-2 K-4

T درجه حرارت مطلق به درجه K: Kelvin (1 = 1 درجه کلوین درجه سلسیوس(

)( 4TeA

Page 24: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

24

Target signature variability and false alarm rate

و با توجه به اينكه قسمت هائي از هدف داراي گرميو سردي متفاوتي هستند ،

از انجائيكه ميزان تابش دريافتي از خورشيد در طولروز متفاوت است و انعكاس آن نيز از سطح هدف

متفاوت ميباشد ،

بنا براين چون اين قسمت ها ممكن است زود ترخنك شوند ، ممكن است داراي كنتراست معكوسي

بشوند ،

Page 25: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

25

Target signature variability and false alarm rate

Note at the tiers intensity in different hours of a day.

Page 26: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

26

Target signature variability and false alarm rate

یاد آوری: الستیک خود رو در اثر تابش خورشید زود تر از رینگ گرم میشود و پس از

غروب خورشید , زود تر نیز خنک میشود.گرمای زیاد با سطح روشنائی زیاد نشان داده شده است.

Page 27: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

27

The Hough Transform in ATR

از آنجائي كه بسياري از اشيا ساخت انسان ويژه گياز تبديل هاف ميتوان براي شناسائي خط مستقيم دارند ،

آنها استفاده كرد.

بعنوان مثال ، از تبديل هاف ميتوان براي شناسائي مسيرحرکت موشك در فضا استفاده نمود.

از آنجائي كه مسير موشك در فواصل زمانی معینی ، تقريبامستقيم است ، در حضور نويز زياد تبديل هاف گزينه بسيار

مناسبي براي تعيين مسير حركت آن است.

از طريق تفاضل گيري بر روي فريم هاي متوالي ، هدف بامقدار نويز زياد را میتوان مشخص نمود.

Page 28: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

28

The Hough Transform in ATR

دقت كنيد كه شكل اين سفينه داراي خطوط مستقيم بسياري است.

تصوير برداري SAR: Synthetic Aperture Radarاگر سفينه با سيستم رادار شود ،

خواهد بود.dropoutsاين تصوير شامل لكه ها و قطعات جدا از هم ( )

Page 29: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

29

The Hough Transform in ATR

البته اگر سفينه در حال حركت باشد ، دنباله مسير طوالني تري را تشكيل ميدهد و تشخيص مسير حركت آن با تبديل هاف که بر روی تصاویر ضبط شده از موشک در موقعیت های متفاوت

و تبدیل مجموعه تصاویر به یک نسخه پانا روما ، امکان پذیر خواهد بود.

تشیخص مسیر حرکت در مورد موشك ها ، توسط سیستم های نيز امكان پذير است.IRتشخیص مبتنی بر سیستم

Page 30: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

30

End of Chapter

Page 31: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

31

Page 32: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

SAR: Synthetic Aperture Radar Imaging Concept, image taken from Airplane

32

Page 33: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

33

SAR: Synthetic Aperture Radar Imaging Concept, image taken from Airplane

Page 34: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

• What is Synthetic Aperture Radar?• Environmental monitoring, earth-resource mapping, and military systems require

broad-area imaging at high resolutions. Many times the imagery must be acquired in inclement weather or during night as well as day. Synthetic Aperture Radar (SAR) provides such a capability. SAR systems take advantage of the long-range propagation characteristics of radar signals and the complex information processing capability of modern digital electronics to provide high resolution imagery. Synthetic aperture radar complements photographic and other optical imaging capabilities because of the minimum constraints on time-of-day and atmospheric conditions and because of the unique responses of terrain and cultural targets to radar frequencies.

• Synthetic aperture radar technology has provided terrain structural information to geologists for mineral exploration, oil spill boundaries on water to environmentalists, sea state and ice hazard maps to navigators, and reconnaissance and targeting information to military operations. There are many other applications or potential applications. Some of these, particularly civilian, have not yet been adequately explored because lower cost electronics are just beginning to make SAR technology economical for smaller scale uses.

34

Page 35: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

35

What is Synthetic Aperture Radar?

How does Synthetic Aperture Radar work?A detailed description of the theory of operation of SAR is complex and beyond the scope of this document. Instead, this page is intended to give the reader an intuitive feel for how synthetic aperture radar works.

Consider an airborne SAR imaging perpendicular to the aircraft velocity as shown in the figure below. Typically, SARs produce a two-dimensional (2-D) image. One dimension in the image is called range (or cross track) and is a measure of the "line-of-sight" distance from the radar to the target. Range measurement and resolution are achieved in synthetic aperture radar in the same manner as most other radars: Range is determined by precisely measuring the time from transmission of a pulse to receiving the echo from a target and, in the simplest SAR, range resolution is determined by the transmitted pulse width, i.e. narrow pulses yield fine range resolution.

Page 36: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

• Synthetic Aperture Radar Imaging Concept•

The other dimension is called azimuth (or along track) and is perpendicular to range. It is the ability of SAR to produce relatively fine azimuth resolution that differentiates it from other radars. To obtain fine azimuth resolution, a physically large antenna is needed to focus the transmitted and received energy into a sharp beam. The sharpness of the beam defines the azimuth resolution. Similarly, optical systems, such as telescopes, require large apertures (mirrors or lenses which are analogous to the radar antenna) to obtain fine imaging resolution. Since SARs are much lower in frequency than optical systems, even moderate SAR resolutions require an antenna physically larger than can be practically carried by an airborne platform: antenna lengths several hundred meters long are often required. However, an airborne radar could collect data while flying this distance and then process the data as if it came from a physically long antenna. The distance the aircraft flies in synthesizing the antenna is known as the synthetic aperture. A narrow synthetic beamwidth results from the relatively long synthetic aperture, which yields finer resolution than is possible from a smaller physical antenna.

36

Page 37: 1 Automatic Target Recognition (ATR) From the book: Machine Vision Authors: W.E.Snyder & H.Qi Cambridge University Press, 2004

• Synthetic Aperture Radar Imaging Concept•

Achieving fine azimuth resolution may also be described from a doppler processing viewpoint. A target's position along the flight path determines the doppler frequency of its echoes: Targets ahead of the aircraft produce a positive doppler offset; targets behind the aircraft produce a negative offset. As the aircraft flies a distance (the synthetic aperture), echoes are resolved into a number of doppler frequencies. The target's doppler frequency determines its azimuth position.

• While this section attempts to provide an intuitive understanding, SARs are not as simple as described above. Transmitting short pulses to provide range resolution is generally not practical. Typically, longer pulses with wide-bandwidth modulation are transmitted which complicates the range processing but decreases the peak power requirements on the transmitter. For even moderate azimuth resolutions, a target's range to each location on the synthetic aperture changes along the synthetic aperture. The energy reflected from the target must be "mathematically focused" to compensate for the range dependence across the aperture prior to image formation. Additionally, for fine-resolution systems, the range and azimuth processing is coupled (dependent on each other) which also greatly increases the computational processing.

37