Title :
Research on the Rice Counting Method Based on Connected Component Labeling
Author :
Huang Jing ; Liu Peiyuan ; Chen Hanwei
Author_Institution :
Coll. of Inf., Zhejiang Univ. of Sci. & Technol., Hangzhou, China
Abstract :
Rice counting is essential to modern agricultural production sectors, counting accuracy directly impact assessment of the merits of rice. In order to solve the problems of time-consuming and labor-intensive and low-precision existing in traditional manual counting and outline counting, this paper uses image processing technology to count rice. Considering the rice overlapping segmentation is not ideal, this paper uses dynamic threshold method to binary the image and then extracts and labels connected domain. Finally, this method gets the number of rice through processing area of each connected domain.
Keywords :
crops; image segmentation; median filters; connected component labeling; connected domain extraction; connected domain labeling; dynamic threshold method; image processing technology; median filtering; rice counting method; rice overlapping segmentation; Automation; Mechatronics; Connected domain; Counting; Dynamic threshold; Median filter;
Conference_Titel :
Measuring Technology and Mechatronics Automation (ICMTMA), 2014 Sixth International Conference on
Conference_Location :
Zhangjiajie
Print_ISBN :
978-1-4799-3434-8
DOI :
10.1109/ICMTMA.2014.133