MODEL BASED 3D POINT CLOUD SEGMENTATION FOR AUTOMATED SELECTIVE BROCCOLI HARVESTING

Montes, Hector, Duckett, Tom and Cielniak, Grzegorz (2019) MODEL BASED 3D POINT CLOUD SEGMENTATION FOR AUTOMATED SELECTIVE BROCCOLI HARVESTING. In: Smart Industry Workshop 2019, 9-11 January 2019, Nottingham Trent University, Nottingham, UK.

Documents
SmartIndustry2019.pdf
[img]
[Download]
[img] PDF
SmartIndustry2019.pdf - Image

10MB
Item Type:Conference or Workshop contribution (Poster)
Item Status:Live Archive

Abstract

Segmentation of 3D objects in cluttered scenes is a highly relevant problem. Given a 3D point cloud produced by a depth sensor, the goal is to separate objects of interest in the foreground from other elements in the background. We research 3D imaging methods to accurately segment and identify broccoli plants in the field. The ability to separate parts into different sets of sensor readings is an important task towards this goal. Our research is focused on the broccoli head segmentation problem as a first step towards size estimation of each broccoli crop in order to establish whether or not it is suitable for cutting.

Keywords:3D segmentation, broccoli segmentation
Subjects:G Mathematical and Computer Sciences > G400 Computer Science
Divisions:College of Science > School of Computer Science
ID Code:39207
Deposited On:23 Dec 2019 11:41

Repository Staff Only: item control page