christopher parrish ece533 project december 2006
DESCRIPTION
The Hough Transform for Vertical Object Recognition in 3D Images Generated from Airborne Lidar Data. Christopher Parrish ECE533 Project December 2006. Airborne Lidar. GPS Reference. Station . Airport Obstruction Surveying. Lidar Point Cloud. - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Christopher Parrish ECE533 Project December 2006](https://reader033.vdocument.in/reader033/viewer/2022050809/56815cf7550346895dcaf98c/html5/thumbnails/1.jpg)
The Hough Transform for Vertical Object Recognition in 3D Images
Generated from Airborne Lidar Data
Christopher ParrishECE533 ProjectDecember 2006
![Page 2: Christopher Parrish ECE533 Project December 2006](https://reader033.vdocument.in/reader033/viewer/2022050809/56815cf7550346895dcaf98c/html5/thumbnails/2.jpg)
GPS Reference Station
Airborne Lidar
Airport Obstruction
Surveying
![Page 3: Christopher Parrish ECE533 Project December 2006](https://reader033.vdocument.in/reader033/viewer/2022050809/56815cf7550346895dcaf98c/html5/thumbnails/3.jpg)
Lidar Point Cloud
Voxelize
3D Grayscale Intensity Image
3D Sobeloperator
3D Grayscale Edge Image
Threshold segmentation
3D Binary Edge Image
Hough Transform to identify vertical cylinders
Vertical objects of interest
Hough transform- based approach for detecting vertical objects of cylindrical shape:
![Page 4: Christopher Parrish ECE533 Project December 2006](https://reader033.vdocument.in/reader033/viewer/2022050809/56815cf7550346895dcaf98c/html5/thumbnails/4.jpg)
3D Grayscale Image2D Color Image Laser Point Cloud
![Page 5: Christopher Parrish ECE533 Project December 2006](https://reader033.vdocument.in/reader033/viewer/2022050809/56815cf7550346895dcaf98c/html5/thumbnails/5.jpg)
T
Tzyx z
fyf
xfGGG
f
222zyx GGGf f
242000242
,484000484
,242000242
:xG
202404202
,404808404
,202404202
:yG
242484242
,000000000
,242484242
:zG
otherwise0 if 1
),,(Tf(x,y,z)
zyxg
Gradient of a 3D image, f(x,y,z):
Magnitude of the gradient:
3D Sobel operator (three 3x3x3 filters expressed here as sets of three 2D matrices)
Thresholded (binary) edge image
Computing Binary Edge Image:
![Page 6: Christopher Parrish ECE533 Project December 2006](https://reader033.vdocument.in/reader033/viewer/2022050809/56815cf7550346895dcaf98c/html5/thumbnails/6.jpg)
3D Binary Edge Images
![Page 7: Christopher Parrish ECE533 Project December 2006](https://reader033.vdocument.in/reader033/viewer/2022050809/56815cf7550346895dcaf98c/html5/thumbnails/7.jpg)
HT Cylinder Detection Algorithm:
Input = 3D binary edge imageQuantize 3D parameter space. Initialize all accumulator cells to zero. For each nonzero voxel in 3D binary edge image,
step through all values of s and t. At each location:
Solve for r Round r to its nearest accumulator cell value Increment counter for that (s,t,r) accumulator cell.
Find entry in 3D accumulator array with highest # of votes.
Assume cylinders are vertical (axes parallel to mapping frame Z axis) => # of parameters reduced from 5 to 3. Representation: (X-s)2+(Y-t)2 = r2
![Page 8: Christopher Parrish ECE533 Project December 2006](https://reader033.vdocument.in/reader033/viewer/2022050809/56815cf7550346895dcaf98c/html5/thumbnails/8.jpg)
Cylinders Detected Using Hough Transform:
![Page 9: Christopher Parrish ECE533 Project December 2006](https://reader033.vdocument.in/reader033/viewer/2022050809/56815cf7550346895dcaf98c/html5/thumbnails/9.jpg)
Comparison of radii & axes locations of HT-detected cylinders with
field-surveyed data: