中国农业科学院农业信息研究所 中国农学会
Agricultural Information Institute of CAAS China Association of Agricultural Science Societies
En
Ch
中国农业科学院农业信息研究所 中国农学会 主办

AI检索

订阅

返回顶部
当前位置: 首页 > 文章 > 基于图像测量的毛竹笋高生长在线监测 林业工程学报 2021 (4) 134-139
Position: Home > Articles > On-line monitoring of bamboo shoot growth height based on image measurement Journal of Forestry Engineering 2021 (4) 134-139

基于图像测量的毛竹笋高生长在线监测

作  者：
贾新宇;江朝晖;李娟;高健
单  位：
国际竹藤中心;安徽农业大学信息与计算机学院
关键词：
毛竹笋;高生长;图像测量;图像分割;实时监测
摘  要：
毛竹笋的高生长信息被视为反映毛竹长势和产量的重要指标,是毛竹生长发育研究中的重要内容。针对现有人工观测方法费时费力,难以实现大范围内实时监测的问题,提出了一种基于图像的毛竹笋高生长监测系统。首先通过架设林间监控摄像头实时获取毛竹笋图像,然后利用几何角度原理,计算图像中毛竹笋的像素数与摄像机旋转角度对应的线性映射关系,进而获取毛竹笋实时高生长数据。对于林间毛竹笋目标分割提取的难题,提出一种结合正弦余弦优化的人工蜂群和模糊局部C均值聚类的图像分割算法。首先在模糊聚类的基础上引入局部空间信息,然后利用正弦余弦优化的人工蜂群算法对聚类算法的初始聚类中心进行寻优,最后再对图像进行二次分割。该方法相比于PSO、ABC、MFO等算法,具有更加优异的寻优性能;相比于FCM和ABC-FCM,也具有更强的抗噪性能和更高的分割精度。将毛竹笋实际高生长测量结果与人工测量值相比,平均误差率仅为4.17%。实验结果表明,此方法能够有效实现毛竹笋高生长的准确实时监测。
译  名：
On-line monitoring of bamboo shoot growth height based on image measurement
作  者：
JIA Xinyu;JIANG Zhaohui;LI Juan;GAO Jian;College of Information and Computer Science, Anhui Agricultural University;International Center for Bamboo and Rattan;
关键词：
moso shoots;;growth height;;image measurement;;image segmentation;;real-time monitoring
摘  要：
The growth height information of moso bamboo shoot is an important index to reflect the growth and yield of bamboo, so it becomes an important subject of the growth and development research of moso bamboo. Aiming at solving the problem that the existing manual observation methods are time-consuming and laborious, which is difficult to monitor in real time in a large range, an image-based growth height monitoring system for moso bamboo shoot was proposed. The bamboo shoot image was acquired in real time by setting up a monitoring camera in the forest, with the consideration of the segmentation and extraction of bamboo shoot under a complex environment. Then the linear mapping relationship between the pixel number of bamboo shoot in the image and the camera rotation angle was calculated using the geometric angle principle, and the real-time growth height data of bamboo shoot was obtained. For the problem of segmentation and extraction of wild moso bamboo shoots from forest, an image segmentation algorithm combining artificial bee colony(ABC) optimization with fuzzy local c-means clustering was proposed. Firstly, in order to reduce the influence of noise points on segmentation performance, the local spatial information was introduced on the basis of fuzzy clustering, and then the initial clustering center of the clustering algorithm was optimized by sines and cosines artificial swarm optimization algorithm. Finally, the image is segmented twice to obtain the ideal segmentation target of bamboo shoot. Compared with the particle swarm optimization(PSO), ABC and moth-flame optimization(MFO), this method has better optimization performance in initial cluster center optimization. Compared with fuzzy c means(FCM) and ABC-based FCM, the segmentation of the real images of moso bamboo shoots in the forest also had stronger anti-noise performance and higher segmentation accuracy. The average error rate was only 4.17% when comparing the actual growth height measurement results with the manual measurement results. The experimental results showed that this method can effectively realize the accurate and real-time monitoring of bamboo shoot growth height.

相似文章

导出引用
计量
文章访问数: 8
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊

友情链接Friendly link
农业信息研究所Agricultural information institute 农业智能知识服务平台Agricultural intelligent knowledge service platform 农业学术服务平台Agricultural Knowledge Service System 中国知网CNKI 维普网VIP 万方WANFANG DATA
联系我们Contact us

电话Tell：010-82109652-808

邮箱E-mail：agrijournal@caas.cn

服务Service
关于我们About us 意见建议Opinions and suggest
版权说明Copyright
单位：中国农业科学院农业信息研究所 中国农学会 © 2017中国农业科学院农业信息研究所 ICP备案号：京ICP备09089781号-46