推荐星级:
- 1
- 2
- 3
- 4
- 5
基于机器视觉的播种精度检测技术
资料介绍
文档为基于机器视觉的播种精度检测技术详解文档,是一份不错的参考资料,感兴趣的可以下载看看,,,,,,,,,,,,,
部分文件列表
文件名 | 大小 |
基于机器视觉的播种精度检测技术.pdf | 281K |
部分页面预览
(完整内容请下载后查看)Seeding Precision Test Based on Machine Vision
Lin Jiachun1 Liwei2
College of Engineering, China Agricultural University
Abstract
Seeding uniformity is a crucial index to seeding performance and quality of a grain seeder. A
testing approach for seeding precision was developed in this paper, which is an integrated technology of
machine vision, pattern recognition, and automatic control. The new method realizes a precise and
reliable test of seeding precision, and an automatic control of tester operation. A machine vision based
test-bed was developed for performance tests of grain seeders, such as a precision planter, a drill
planter, a hill-drop planter, and a corresponding software package was compiled to capture the images
of the deposited seeds, to segment the seeds from the background of an image, and to calculate the
distances between seeds after precision seeding, the number of seeds per length after drill seeding, and
the distance between hills and number of seeds per hill after hill-drop seeding. The date obtained above
was then used to estimate the performance of a seeder by the statistical calculations. At the end of the
test, a normative testing report was printed out. A special illumination system was constructed to light
the scenes under the cameras. Three monochrome cameras, one per row, were connected to a color
frame grabber board to accomplish image captures. An special designed image splicing algorithm was
enable to eliminate overlapped area of two adjacent images in sequence. The test-bed has proved
reliability and accuracy from many practical tests.
Keywords: Seeder performance test, Machine vision, Sequential image, Image splicing
1. Introduction
Performance of a seeder has remarkable influences on cost and yield of agricultural products and
especially its seeding uniformity is a crucial index in estimation of a seeder’s quality. In order to
quantify a planter performance, a wide variety of approaches have been developed with regard to plant
spacing[1~11]. Performance measurements, such as the distance between plants in the field, have been
used in some tests, the distance between seeds on a grease covered belt test stand in the other tests and
the distance between seeds planted into soil in a few tests.
A planter test at present involves the techniques of optoelectronics, piezoelectric method, fast
imaging, and computer vision. Lan, Kocher and Smith developed an opto-electronic sensor system for
laboratory measurement of planter seed spacing with small seeds[11]. The core of the hardware in this
opto-electronic system is a shutter that consists of 24 pairs of near infrared (NIR) LEDs and
phototransistors. Shi Zhixing developed a seed detection sensor, which consists of a red light diode
(RLD) and a silicon photoelectric cell, used for planter performance test. The light from the diode was
reflected by two parallel placed mirrors to form light grid, which would be shaded when a seed fell [9]
.
1
Support by Agriculture Seed Project, (200KY001)
1
全部评论(0)