Assessment of a ground-based weed mapping system in maize

Weed spatial distribution was measured with a ground-based weed mapping system. The system consists of three different components: i) weed detection by optoelectronic sensors; ii) weed geopositioning by DGPS receiver; and iii) a data acquisition and processing system. Three optoelectronic modules we...

Descripción completa

Detalles Bibliográficos
Autores: Andújar, Dionisio, Ribeiro Seijas, Ángela, Fernández-Quintanilla, César, Dorado, José
Tipo de recurso: artículo
Estado:Versión aceptada para publicación
Fecha de publicación:2009
País:España
Institución:Consejo Superior de Investigaciones Científicas (CSIC)
Repositorio:DIGITAL.CSIC. Repositorio Institucional del CSIC
OAI Identifier:oai:digital.csic.es:10261/243586
Acceso en línea:http://hdl.handle.net/10261/243586
Access Level:acceso abierto
Palabra clave:Weed detection
Optoelectronic sensors
Sampling resolution
Descripción
Sumario:Weed spatial distribution was measured with a ground-based weed mapping system. The system consists of three different components: i) weed detection by optoelectronic sensors; ii) weed geopositioning by DGPS receiver; and iii) a data acquisition and processing system. Three optoelectronic modules were mounted on the front of a tractor at a 0.75 cm distance between them and at 60 cm height above ground level. Consequently, the system was able to explore a 2.25 m band [isn't that more than one row? This is with all three sensors?], corresponding to three crop rows, detecting the vegetation present in the middle of the inter-rows maize area. The working capacity of the system was higher than 1 ha h-1. The weed mapping system was evaluated in three maize fields during spring, when the crop was [in/at] 4-to-6 leaves stage. In order to verify the system with highly reliable data, digital images were obtained in random geo-referenced points distributed throughout the three fields. Three experienced observers rated these images for weed presence/absence, using a presence threshold of 15% weed cover. The comparison between the data obtained with the ground-weed mapping system and from the values derived from visual assessments of the digital images indicated a good agreement (84% on average) between the two sets of data. The comparison among the results obtained with various simulated distances between sensors (from 1.5 m to 6.0 m) indicated that the ground-based system could construct weed maps accurately using a distance between optic sensors of 4.5 m.