论文信息 - Tracking of row structure in three crops using image analysis

Tracking of row structure in three crops using image analysis

Abstract This paper is concerned with visual sensing for an autonomous crop protection robot, in particular the problem of finding guidance information from crop row structure. A robust method for finding crop rows in images has been presented in the author's previous work and is briefly re-introduced here. This is based on the Hough transform but, unlike previous methods, integrates information over a number of crop rows making the technique very tolerant to such problems as missing plants and weeds. Experiments are reported where the method is tested on three crops; cauliflowers, sugar beet, and widely spaced double rows of wheat. When compared with a human assessment of the row positions in images, typical errors were 18 mm of lateral offset and 1° of angle.

J. A. Marchant | J. Marchant

[1] R. Y. Tsai,et al. An Efficient and Accurate Camera Calibration Technique for 3D Machine Vision , 1986, CVPR 1986.

[2] R. D. Tillett,et al. Image analysis for agricultural processes: a review of potential opportunities , 1991 .

[3] J. A. Marchant,et al. SEGMENTATION OF PLANTS AND WEEDS USING INFRARED IMAGES , 1996 .

[4] S. W. Searcy,et al. Distributed control for vision based tractor guidance , 1989 .

[5] Gongzhu Hu,et al. Image processing for path-finding in agricultural field operations. , 1985 .

[6] N. D. Tillett,et al. Navigation and control of an autonomous horticultural robot , 1996 .

[7] John A. Marchant,et al. Real-Time Tracking of Plant Rows Using a Hough Transform , 1995, Real Time Imaging.

[8] J. V. Stafford,et al. In-field location using GPS for spatially variable field operations , 1994 .

[9] H. J. Olsen. Determination of row position in small-grain crops by analysis of video images , 1995 .

[10] Neil H. Brown,et al. Image analysis for vision-based agricultural vehicle guidance , 1991, Other Conferences.