穨封面.PDF

Size: px

Start display at page:

Download "穨封面.PDF"

完爷东
7 years ago
Views:

2 ( ) ( \/ ) ( ) ( ) ( ) ( ) ( ) ( ) : :

3 ( ) :

4 ABSTRACT The interpretation of building edges is complex and difficult in the automated reconstruction of buildings. In a semi-automated approach, the interpretation may be done manually, then at primary reconstructive period the process by computer at later periods not only maintains the precision of reconstruction but also reduces the unnecessary operation and loading. This investigation uses manually digitization for rough corner locations of the rectangular man-made buildings in an image of a stereopair. From the collection of those rough corner points, a method of being developed could result in reasonable roof lines, called CONTOURS, by using the property of orthogonalization as constraints. Then, the edges in a single image may be detected in a local buffer for every contour. Along the edges point-by-point, image matching with its conjugate image could determine disparities. Finally, from candidates of one contour we select the edge that has the most supports according to the height relation between the edges. Thus, the intersections of edges are the corners of the building proof.by space intersection, the 3D model of the building has reconstructed.. Key words: Semi-automatic, Manual Interpretation, Local Edge Detection, Image Matching

9 I

10 II Hough Hough Hough CLR CLR

11 III

12 electric map Urban Planning City Model [Kim & Muller, 1998; Sahar & Krupnik, 1999] 1

13 Gabet [Gabet et al., 1997] Digital Elevation ModelDEM ETH AMOBEAutomation of digital terrain Model man-made OBject Extraction from aerial omages ARUBAAutomation Reconstruction of Building from Aerial images [Henricsson, 1996Gruen, 1997] TOBAGO TOpogy Builder for the Automated Generation of Object from 3-D point clouds ARUBA 2

14 TOBAGO Frere [Frere et al., 1997] 1 Canny Fisher [Fisher et al., 1997] semantics Scanning Laser Altitude and Reflectance Sensor, ScaLARS Digital Surface Model, DSM [Haala, 1997] 80~ 90 [Sahar et al., 1999] [ 1998] 3

15 [Huertas & Nevatia, 1988Mohan & Nevatia, 1989] Global Operation 4

16 1.1 5

17 6

18 A 2-1 B 2-2 A 1 2 7

19 L HOUGH 8

20 (a) (b) (c) 2-2 (a) (b) (c) 3 = (a) 9

21 0 < 1 2 < <0 2-3(b) ct =0 2-3(c) 1 L T 2 L L 2-4(a) LT 4 T T 2-4(b) 4 T 1 2 (a)l (b)l (c)t 10

22 L L T L 2 T 11

23 Derivative Operators Multi-Resolution Edge Detection MEDTDeok et al., A. P=1 3 3 P=1 12

24 P=1 0 B. 2 max +1 2 min+1 B-1. A 2 max +1 Pmax B-2. B-1 2 max-1 +1 Pmax-1 B-3. B-2 2 min+1 Pmin max- min max- min +1 max- min +1 DROG Derivation of Gaussian[Deok et al., 1995] Convolution DROG H R H C [ ga( x, s) ga( y, t)] H 2.1 R ( x, y) = x 2.2 xga( x, s) ga( y, t) H R ( x, y) = 2 s 13

25 [ ga( x, s) ga( y, t)] H ( x, y) 2.3 C = y 2.4 ga(x,s)ga(y,t) s t 2-5 H C ( x, y) = yga( x, s) ga( y, t) 2 t SNN Filter r(i,j) 1 0 DROG σ=σ 0 +(p max -p min +1) σ r(i,j)<γ p=1 r(i,j)<γ p max p max -p min +1 p max -p min +2 DROG σ=σ 0 + σ DROG σ=σ 0 r(i,j)<γ p min

26 Local Maximum Trace[ & 1997] : A. Seed Points A-1. E1 A-2. B. A 3 3 E2 Branch Point C. E3 15

27 D E1 E3 HOUGH Hough [Hough ] Hough Feature Point 2.5 Parameter Space Cluster Detection Hough 16

28 [ 1995] = x cos+ y sin 2.5 x ~180 D D A Hough 17

29 2-6 ABC Hough [ 1998] Hough M N : A Hough r,c Hough x y 2-8 Y X 2-8 Hough A A M180 0 / ND/ Ai, j 0 i=1,2,.m j= 1,2, N 1 L 18

30 2.8 Spread Leavers, > tan 1 ( / L ) 2.8 i = i i=1,2,.m 2.9 i 2.10 j j j = x cos i + y sin i j = j / A(i,j)= A(i,j) C Adaptive Cluster Detection [Risse, 1989] : C1. k l C2. P l = x cos k + y sin k P L C3.L B B 19

31 D Hough : ax+by=c 2.13 a b c 2.13 c d 1.5 x y C1~C3 () g Max 2-9 Hough D1 20

32 2-9 Hough 21

34 Matching Primitives [Hannah, 1988] CLR [ 1999] Target Window Search Window Normalized Cross Correlation, NCC (Moving Window) CLR 3.2 Operate Win-dow Correlation Coefficient [( ) ( )] i, j Ws Ws Wt Wt ( W W ) ( W W ) NCC = 2 1/ i, j s s t t 23

35 Ws W t iline jsample CLR 3-2 CLR [ 1999] CLR 3-4 CLR NCC 24

36 CLR 3-4 CLR 25

37 L T [ 1999] A Hough

38 3-5 B A P L P S P L P S P 0 L1L2 P 3-6 (L p, S p,0)

39 3-6P L P S P 0D1 L1 P D2 L2 P L1 L2 m 1 m 2 H c 1 c 2 L1 P c 1 L2 P c 2 c 1 c 2 Support 1 L 2 T L T 28

40 29 L T T i = 1~2( ) xy f XY M = + = = + = i i i i i i i i i i W V f y F W U f x F = Li Li Li i i i i Z Z Y Y X X M W V U

41 60 1/ µm a39-16b39-15 (a) (b) 4-1 (a) (b)

42 4-2-1 Bottom-Up [Tudor & Sugarbaker 1993Chen & Lee1993] A B : 31

43 1 E140 2 E260 3 E320 C 8 1 D Target Window 5Line 11Sample (a) (b) 4-2 a b 32

44 (a) (b) 4-3 a b (a) (b) 4-4 a b 33

45 (b) 4-6 5th 4th 1st 2nd 3rd (a) (b) 4-5 (a) (b) 34

46 4-3-2 MEDT 4-7(a) (b) HOUGH 4-7(c) 35

47 (a) (b) (c) HOUGH 36

48 a b 4-8 a b 4-9 ( ) 37

49 #1 #4 #3 # (a)(b) (m) (m) E N H E N H #1 #2 #3 #4 4-1 (a) (m) deltae deltan DeltaH 2 #1 #2 #3 #4 RMSError RMSE 4-1 (b) 38

51 T Ambiguity 40

53 4-13 HOUGH

54 43

55 4-15 ( ) (a)(b)

56 #11 #10 #9 #8 #7 #6 #5 #4 #3 #2 #1 #12 #14 #13 #16 #15 #18 #17 #20 #19 #22 # (a) 4-16 (b) 45

57 4.2(b) (m) (m) E N H E N H #1 #2 #3 #4 #5 #6 #7 #8 #9 #10 #11 #12 #13 #14 #15 #16 #17 #18 #19 #20 #21 # (a) 46

58 (m) deltae deltan deltah #1 #2 #3 #4 #5 #6 #7 #8 #9 #10 #11 #12 #13 #14 #15 #16 #17 #18 #19 #20 #21 #22 RMSE 4-2 (b)

59 H

63 4-19 HOUGH

64 53

65 54

66 4-21 ( ) 55

67 #12 #11 #10 #9 #8 #7 #1 #2 #3 #4 #5 # (a) 56

68 4-22 (b) (a)(b)

69 (b) (m) E N H E N H #1 #2 #3 #4 #5 #6 #7 #8 #9 #10 #11 # (a) 58

70 (m) deltae deltan deltah RMSE 4-3 (b)

73 " 1996 " Hough " 1995 " 1999 " 1999 " 1999 " 1996 " 1998 " Vol. 23No. 4pp " Hough " pp Deok, J. P., Nam, K. M. and Park, R. H., "Multi-resolution edge detection techniques", Pattern Recognition, Vol. 28, No. 2, pp ,1995 Fisher, A., Kolbe, T. H., Lang, F., "Intergration of 2D and 3D Reasoning for Building Reconstruction Using a Generic Hierarchical Model", Semantic Modeling for the Acquisition of Topographic 62

74 Information from images and Maps, Birkhauser Verlag, Berlin, pp , 1997 Frere, D., Hendrickx, M., Vandekerckhove. J., Moons, T., and Gool, L. V., "On the Reconstruction of Urban House Roofs from Aerial Images, Automatic Extraction of Man-made Objects from Aerial and Space Images(II), Birkhauser Verlag, Berlin, pp , 1997 Gabet, L., Giraudon, G., and Renouard, L., "Automatic Generation of High Resolution Urban Zone Digital Elevation Models", ISPRS Journal of Photogrammetry and Remote Sensing, Vol. 52, No. 4, pp , 1997 Gruen, A., and Dah, H., "TOBAGO- A Topology Builder for the Automated Generation of Building Model", Automatic Extraction of Man-made Objects from Aerial and Space Images(II), Birkhauser Verlag, Berlin, pp , 1997 Halla, N., and Brenner, C., "Interpretation of Urban Surface Models Using 2D Building Information", Automatic Extraction of Man-Made Objects from Aerial and Space Images(II), Birkhauser Verlag, Berlin, pp ,1997 Hannah, M., "Digital Stereo Image Matching Techniques", ISPRS 16 th Congress Kyoto, Com. III, pp , 1988 Henricsson, O., Bignone, F., Willuhn, W., Aef, F., Kuebler, O., Baltsavisa, E., Mason, S., and Greun, A., "PROJECT AMBOE: Strategies, Current Status and Future Work", ISPRS Congress, Commission III, Vol. 31, Part B3, pp , 1996 Hough, P.V.C., "Methods and Means for Recognising Complex Patterns", U.S. patent No , 1962 Huertas, A., and Nevatia, R., "Detecting Buildings in Aerial Images", Computer Vision, Graphics, and Image Processing, Vol. 41, pp ,

75 Kirn, T., and Muller, J.P., "A Technique for 3D Building Reconstruction", Photogrammetric Engineering and Remote Sensing, Vol. 64, No. 9, pp , 1998 Leavers, V.F., "Shape Detection in Computer Vision Using the Hough Transform", Springer-Verlag London Limited, 1992 Mohan, R., and Nevatia, R., "Using Perceptual Organization to Extract 3-D Structures", IEEE Trans. On Pattern Analysis and Machine Intelligence, Vol. 11, No. 11, pp , 1989 Sahar, L., and Krupnik, A., "Semiautomatic Extraction of Building Outlines from Large-Scale Aerial Images", Photogrammetric Engineering and Remote Sensing, Vol. 65, No. 4, pp ,

2 3. 1,,,.,., CAD,,,. : 1) :, 1,,. ; 2) :,, ; 3) :,; 4) : Fig. 1 Flowchart of generation and application of 3D2digital2building 2 :.. 3 : 1) :,

2 3. 1,,,.,., CAD,,,. : 1) :, 1,,. ; 2) :,, ; 3) :,; 4) : Fig. 1 Flowchart of generation and application of 3D2digital2building 2 :.. 3 : 1) :, 3 1 Vol. 3. 1 2008 2 CAA I Transactions on Intelligent Systems Feb. 2008, (,210093) :.,; 3., 3. :; ; ; ; : TP391 :A :167324785 (2008) 0120001208 A system f or automatic generation of 3D building models

穨封面.PDF

穨封面.PDF