Using non-contact sensing techniques and facilities to gain plant three-dimensional (3D) information can satisfy different kinds of needs in digital and intelligent agricultural production. 3D information could be used for vegetative modeling and monitoring or plant production management and operation. A plant 3D information detecting technology based on the movement of single camera was proposed. The method was divided into three steps. Firstly, calculate the camera matrix in different spatial positions. Presume that world coordinate was coincided with camera coordinate in one position. With the known movement parameters of camera, it was able to calculate the camera matrix in another position. Secondly, extract plant information using image processing technology. Then, sample plant areas and pick up matching points. A region matching method based on epipolar constraint was applied to obtain corresponding points between images gathered in different positions. The region matching error metrics was the sum of absolute differences (SAD) of RGB channels. Finally, with the matching points and projective geometry, target plant's 3D information could be calculated. A web camera common on the market was used in experiments to verify this detecting technique. It turned out that while the average distance between camera and plant was less than 1.2m, the detecting error of x and y axis was within -18.9mm~+16.2mm and +7.0~+28.1mm in z direction. While the average distance about 1.5m, the detecting error of x and y axis was within -36.0mm~+5.6mm and +7.0~+59.0mm in z direction. And it also appeared that the closer the distance was the higher the detecting precision would be. Index Terms - information technology, sensing detection, three- dimension (3D), plant
[1]
A. Escolà,et al.
Obtaining the three-dimensional structure of tree orchards from remote 2D terrestrial LIDAR scanning
,
2009
.
[2]
Andrew Zisserman,et al.
Multiple View Geometry in Computer Vision (2nd ed)
,
2003
.
[3]
He Yong,et al.
Research on plant three-dimensional information acquisition method
,
2012
.
[4]
Lie Tang,et al.
Automatic inter-plant spacing sensing at early growth stages using a 3D vision sensor
,
2012
.
[5]
Qamar Uz Zaman,et al.
Quantifying Sources of Error in Ultrasonic Measurements of Citrus Orchards
,
2005
.
[6]
H. Opower.
Multiple view geometry in computer vision
,
2002
.
[7]
Peter Biber,et al.
Plant detection and mapping for agricultural robots using a 3D LIDAR sensor
,
2011,
Robotics Auton. Syst..
[8]
H. Scharr,et al.
A stereo imaging system for measuring structural parameters of plant canopies.
,
2007,
Plant, cell & environment.
[9]
Philippe Lucidarme,et al.
On the use of depth camera for 3D phenotyping of entire plants
,
2012
.