Share Email Print

Proceedings Paper

Pose parameter extraction of corn canopy remote sensing images based on parallel multi-ocular imaging
Author(s): Xin Li; Yan'e Zhang; Jingfu Zhu; Ruijiao Zhao; Minzan Li
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

A 3-dimensional reconstruction model and pose parameter extracting method of parallel multi-ocular image are proposed. The multi-spectral camera is arranged as a rectangle with four channels of R, G, B and NIR. The corn canopy images in field are captured by the camera, and the distance from the camera to the corn canopy is about 0.5m. A novel matching method of feature point is proposed. Channel NIR is taken as the source, and the others are taken as destination. And then the edge of the corn leaf in source image is taken as the source feature points. Feature vector of each point is composed of its 18 directional derivatives. After that, the destination feature point is searched in destination image. First, the local area is estimated where the feature points may lie on. Then, if the intersection angle of two edges, formed by local points and source points, is smaller than a threshold, and the Euclidean distance between feature vectors of local points and source points is minimum among all of them, the local points are thought to match destination feature points, and the feature points pair set is constructed. The edge direction and distance are used as the principle to divide the different area of the image, so that the different leaf regions of canopy image are segmented. The 3-dimensional coordinate of each point in the region can be calculated. From four channel images, at least three 3-dimensional coordinates of each point can attained, and the center of gravity is the more accurately 3-dimensional coordinate. The interpolation is used to reconstruct corn leaf in space, and the pose parameters such as inclination of leaf and so on are estimated.

Paper Details

Date Published: 13 November 2010
PDF: 8 pages
Proc. SPIE 7857, Multispectral, Hyperspectral, and Ultraspectral Remote Sensing Technology, Techniques, and Applications III, 785713 (13 November 2010);
Show Author Affiliations
Xin Li, Heilongjiang Bayi Agriculture Univ. (China)
Yan'e Zhang, China Agricultural Univ. (China)
Jingfu Zhu, Heilongjiang Bayi Agriculture Univ. (China)
Ruijiao Zhao, China Agricultural Univ. (China)
Minzan Li, China Agricultural Univ. (China)

Published in SPIE Proceedings Vol. 7857:
Multispectral, Hyperspectral, and Ultraspectral Remote Sensing Technology, Techniques, and Applications III
Allen M. Larar; Hyo-Sang Chung; Makoto Suzuki, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?