The Sea Journal of the Korean Society of Oceanography Vol. 11, No. 4, pp. 145 151, November 2006 ˆx d š ½k*ÁŸ w www wlœw Study on Intertidal Flat Topography Observation Using Camera Images TAERIM KIM* AND SOECKWANG PARK Ocean system Eng. Dept. Kunsan University, Miryong-Dong, San-68, Kusan, Chonbuk 573-701, Korea ˆ w š ùkü. w w»w w ƒ š w œ t zw mww x w 3 x. 1y d ˆ z d ƒ w ˆ x y œ. ˆ x w l e w d ww x dw, x dw w w ùk ü w» d x l y mw. The water line moving on the intertidal flat during a flood indicates depth contours between low and high water lines. The water lines extracted from the consecutive images are rectified to get the ground coordinates of each depth contour and integrated to provide three dimensional information of intertidal flat topography.gthe tidal flat outside Saemankeum-1 sea dike shows the most obvious changes of tidal flat topography after the construction of sea dikes. This tidal flat topography was observed using digital camera images, and the calculated depths were very similar to in-situ measurement data. Topography changes obtained from two different period data were also examined. Keywords: Intertidal flat, Saemankeum sea dike, Water line, Depth contours, Rectification, Camera images œƒ ü d d ˆ x w yƒ š. k ˆ sx k w x yƒ j ù wù œ sx Á w x yƒ w. w y wü n ƒ jù x» w öeš xw» w. wù ûw w ˆ ¾ y ƒ. ù ˆ w z y ƒ š p d ˆ z w eš. w ˆ x p w wì q *Corresponding author: trkim@kunsan.ac.kr w ˆ x y wš n» w ƒ vw. ù w ˆ x s š w x w d w. s 1.2 km 6 200 m w»» m d w d ƒwš w d w ƒ w. p» w d d ƒw. ˆ x d ˆ tw x ew šy, y» w d d w. ˆ x w x y ù Synthetic aperture radar(sar) w ƒ š ù w y y w ƒ (, 2000).» š» v w ƒw w. šw l e 145
146 ½kÁŸ w w s³,,» ww x dwš w w y yw š. p d d w w»w mw yƒ ƒw. w w x w d» w y q w x w ƒ y š (Holand et al., 1997). Argus( y ù ) vp mw w d ewš d w mw wwš. x w d w Nathaniel and Holman(1997)» swash zone w w s w š w w l e wì Stefan et al.(2003)» d wg e w wš m w x y d/w. Alexander et al.(2004) 4 Argus p zw w ùkù w e q j» w Ranasinghe et al.(2004) 4» w w x y wš e w x y w x w. w Aagaard et al.(2005) st w w y w x d d ww ww. ù ½ (1998) l w w y d» w ˆ d w z x dw w, ½(2003) d l w kt ü w w y w. š x d» ˆ x w s w y y w w(½, 2006). w w d ˆ x wš w. d œ z d ˆ x yƒ wš 1y d ˆ(Fig. 1) 1 y 2 e ew ˆ w x d ww. x d š ùkü w ƒƒ e»w mw e t y ƒw yw. Fig. 2 (a) ssw ƒ x e ùkü š (b) xk e, (c) š xk e ƒƒ ùkü. w w mww š ùkü. w d w ƒƒ s e»w mw Fig. 1. (a) Map of Saemankeum area and (b) a photo scene of Saemankeum-1 sea dike area(image courtesy of KARICO). The area under investigation is marked by a yellow dot and an ellipse, respectively.
ˆx d š 147»(x, y) t w, X, T, Z (x, y) w e t, f c e, X L, Y L, Z L t l e ¾ w. m ij (2) X, Y, Z ƒƒ w z w Mw 3 x 3 ƒƒ w ω X w z, φ Y w z, κ Z w z ùkü ƒƒ w z w. M = cosφcosκ cosωsinκ sinωsinφcosκ + ω κ sin sin cosωsinφcosκ cosωsinκ cosφcosκ sinωsinφsinκ sinωcosκ + cosωsinφsinκ sinφ sinωcosφ cosωcosφ (2) Fig. 2. Example of water depth contours constructed by the movement of flood boundary lines in the case of accretion and erosion. Fig. 3. Geometry of object and image space coordinate systems. z. w»w Fig. 3 t (X L, Y L, Z L)j mw t e k z ƒƒ w z š t j» w t ek m œ(1) (Mikhail and Bethel, 2001). X X L Z Z m 11x+ m 12 y+ m 13 ( = ( L )------------------------------------------------- m 31 x+ m 32 y+ m 33 ( Y Y L Z Z m 21x+ m 22 y+ m 23 ( = ( L )------------------------------------------------- m 31 x+ m 32 y+ m 33 ( (1) X L, Y L, Z L, f c, ω, φ, κ zw» w 7 e w w»(gcp: Ground Control Point) w d. y w 7 GCP d d l, ƒ yw (X L, Y L, Z L, f c, ω, φ, κ) w ü. w w w l w e t(x, y) x dw ƒƒ Z (1) w e t X Y w ƒƒ w et zw š ƒw. w w x d w w. p» d w w ƒ w w ƒ w. wù RGB ƒ w. q ƒ¾ R R w ƒ¾ w p B j ùkù w. ù w» y / Ÿ w ƒ HSV(Hue, Saturation, Value) p wš w w w w» w. ƒ ƒ HES(Human Expert System), mw w. wz w y w ƒ vw. š 1y d ˆ w w ƒ ww ˆ s ˆ w üd w j» 1km ü s (Fig. 4(a)). p ˆ
148 ½kÁŸ Fig. 4. Photos show (a) tidal flat outside the Saemankeum-1 sea dike and (b) series of stakes on the flat. ù w e d w ù» d w x d w (Fig. 4(b)). w d ƒ j š. š w d ƒwš w w d yƒ. 1y ˆ d ƒ¾ ù e ew dw vw e ew dw. d t y e dw mw w. š ˆ x dw» w ƒ d w. w d Ÿ w w š w w. 2005 9 24 z 12 48l z 7 13¾ š 2005 11 18 10 9l z 4 26¾ ww d w. w w 9 24 d 118 cm š 550 cm 432 cm š 11 18 d 36 cm š 631 cm 595 cm. 9 d w e ew dwš 11 d ƒ e ew dw. e šƒ w w ˆ 500 m ¾ d ƒw. d ew l 10»w w»(gcp) d d y w» w» w x d ww. w x x d. Fig. 5» d (1) w»w ww. Fig. 5(a) w š Fig. 5(b) (a)»w w. ùkù Fig. 5. (a) An oblique image of tidal flat during a flood and (b) its rectified image. w ˆ wš l ƒ ˆ w ùkû.
ˆx d š 149 Fig. 6. (a) A scaled rectified image and (b) water depth contours of the area marked by red rectangular box in (a) estimated by camera images on September 24, 2005.»w mw y ƒ w x w ùkù w y w ƒw. 10 l w z x dw w (1) ww ùkù w t(x, Y, Z) w. mww l»¾ ˆx w w. Fig. 6(a) 9 dw w»w ww š Fig. 6(b) Fig. 6(a) ƒx ü w. -7.7 m -6.7 m e w»» w.» cross shore ww w w alongshore w w ùkü. Fig. 7 x dw w w w ƒ Ÿq» el x d¾ w. x d ˆ e Ÿq d»ƒ e w w q¾ sww» w d w e w» w d w. el wš x d w. 11 d e w ˆ x d ww. Fig. 8(a), (b) ƒ ƒ w l w w l w ùkü. e šƒ û ƒ 1y d ˆ w e w s d Fig. 7. Comparison of depths between measured and calculated data. w. Fig. 8 d w ƒƒ e 500 m s d wwš e we ˆ s 1 km w d ƒw. mww e w el mww šwš. ü w wù mww 9 w(fig. 10). Fig. 9 d» d» š»k w x d
150 ½kÁŸ Fig. 8. Water depth contours estimated by images from the camera located on the dike (a) sea side and (b) near land, respectively, on November 18, 2005. Fig. 10. Depth contours (a) in September, (b) in November and (c) depth difference contours between (a) and (b). y w ƒw. w x d mw d w. Fig. 9. Comparison of depths between measured and calculated data. l w w. Fig. 7 ƒ wš d ew w. Fig. 10 9(Fig. 10(a)) 11 (Fig. 10(b)) ˆ dl w w x y(fig. 10(c)) r. w d ƒ 9 d ww ww w. nx w ˆ e w. x» x y w n y w qw». wz» d mw ˆ x 1y d ˆ x 10 w l w z œ w ƒƒ w x t wš mww w. ˆ x w s š ƒ ƒq d ƒ ˆ x x d w w. ƒ w w., e šƒ û ƒ j w ƒ û yw w»., ƒƒ w x œ yƒ j x d w j. ˆ x e w w., ƒƒ w ƒ w š qƒ w
ˆx d š 151 w yw., w ƒ w. ˆ p w yw. w w» w š e w ù y ƒw ƒ. 10d šƒ yw e ew e q/pp l w ƒw w d ww. w ˆ x w w. w ù š œ. w w» w d w š œù y yw. w z w w e vl w dw šw w. w qù w j yw dw w w» w yw qš vw w w ƒ vw. œƒ w, l n š d q s w yƒ. š n e,, n j w öe. p 1y ˆ x w yƒ, w x d w» d ww 1y d ˆ w x p y w œw. w y l(ric) w. x d x(codil) w š wš w w w tw. w w Ì w. šx ½k, Ÿ,, 1998. l w y d» w š. w ww œwz, 10(1): 45 53. ½k, 2003. d l yw w y d. w kwz, 19(2): 99 106. ½k, 2006. e d l w 3 x d, w ww œwz, 18(1): 63 68. x,,, k,,, ½, 2000. k l w w DEM, w kwz, 16(3): 221 233. Aagaard, T., Kroon, A., Andersen, S., Sorensen, R.M., Quartel, S., Vinther, N., 2005. Intertidal beach change during storm conditions; Egmond, The Netherlands, Marine Geology, 218: 65 80. Alexander, P.S. and Holman, R.A., 2004. Quantification of nearshore morphology based on video imaging, Marine Geology, 208: 101 111. Bradley, D. Morris, Mark A. Davidson, David A. Huntley., 2001. Measurements of the response of a coastal inlet using video monitoring techniques, Marine Geology 175: 251 272. Holand, K.T., Holman, R.A., Lippmann, T.C., Stanley, J. and Nathaniel, P., 1997. Practical use of video imagery in nearshore oceanographic field studies. IEEE journal of oceanic engineering, 22: 81V92. Mikhail, E.M. and Bethel, J.S., 2001. Intorduction to modern photogrammetry. John Wiliey & Sons, New York, pp. 90 93. Nathaniel, G.P. and Holman, R.A., 1997. Intertidal beach profile estimation using video images. Marine Geology: 140: 1 24. Ranasinghe, R. Symonds, G., Black, K. Holman, R., 2004. Morphodynamics of intermediate beaches: a video imaging and numerical study, Coastal Engineering 51: 629 655. Stefan, G.J.A., I.L. Turner, Dronkers, T.D.T., Mark Caljouw, Leann Nipius., 2003. A video-based technique for mapping intertidal beach bathymetry. Coastal Engineering 49: 275 289. 2006 4 18 š 2006 9 27 k r: y