Title: Sensors requirements for robust single plant detection in real time
Authors : Seatovic, Dejan
Meiser, Vincent
Scherly, Daniel
Proceedings: Proceedings of the international conference Ag. Eng., AgEng2014
Conference details: AgEng2014 Engineering for Improving Resource Efficiency, Zürich, 6. - 10. Juli 2014
Publisher / Ed. Institution: Zurich
Issue Date: 2014
License (according to publishing contract) : Licence according to publishing contract
Type of review: Not specified
Language : English
Subjects : Plant; Detection; Visual perception; Automation
Subject (DDC) : 621.3: Electrical engineering and electronics
Abstract: This paper considers theoretical prerequisites for the sensor or sensor system that provides data for reliable single plant detection. The goal of the paper is to close the gap between the state of the art sensor technology and the theoretical “ideal” sensor to fulfill one and only purpose: providing single plant detection data in cluttered and disordered environments. Through analysis of artificial leaf forms, the paper determines expectations of the sensor spatial resolution and its maximum noise level. Supported by the results of a theoretical pri-mer on predefined (artificial) shape analysis, an in-depth evaluation compares the results of 3D shape analysis with results obtained through 2D digital image processing. To analyze leaf shapes reliably, a minimum sensor resolution (2D and 3D) is determined by means of statis-tical feature analysis. As a final result of the paper, methods for sensor selection are pro-posed with regard to the specific task of analyzing leaf shape in real time.
Departement: School of Engineering
Organisational Unit: Institute of Mechatronic Systems (IMS)
Publication type: Conference Paper
URI: https://digitalcollection.zhaw.ch/handle/11475/7115
Appears in Collections:Publikationen School of Engineering

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.