Applications of computer vision techniques in viticulture to assess canopy features, cluster morphology and berry size
Computer vision systems are powerful tools to automate inspection tasks in agriculture. Typical target applications of such systems include grading, quality estimation, yield prediction and monitoring, among others. The capabilities of an artificial vision system go beyond the limited human capacity...
| Autores: | , , , , , |
|---|---|
| Formato: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2013 |
| País: | España |
| Recursos: | Universidad de La Rioja (UR) |
| Repositorio: | RIUR. Repositorio Institucional de la Universidad de La Rioja |
| OAI Identifier: | oai:portal.dialnet.es:doc/5bbc69b1b750603269e81fd4 |
| Acesso em linha: | https://investigacion.unirioja.es/documentos/5bbc69b1b750603269e81fd4 |
| Access Level: | acceso abierto |
| Palavra-chave: | Automation Grapevine Image analysis Machine vision Vineyard |
| Resumo: | Computer vision systems are powerful tools to automate inspection tasks in agriculture. Typical target applications of such systems include grading, quality estimation, yield prediction and monitoring, among others. The capabilities of an artificial vision system go beyond the limited human capacity to evaluate long-term processes objectively and provide valuable data to take decisions that will have great influence in later operations. This work explores the application of machine vision techniques in viticulture from several approaches. The first approach is aimed at working outdoors, developing in-field systems capable of assessing the canopy features of the vineyard (Vitis vinifera L.) by taking digital images and applying computer vision systems. The second approach is aimed at analysing cluster morphology using image analysis. Berry number per cluster and cluster weight were estimated using several algorithms of image processing. Lately, machine vision has been used as a tool to automate the measurement of berry size and weight under laboratory conditions. Manual measurement of the canopy features and yield components are tedious and subjective tasks that can be time-consuming and labour demanding. In this regard, by means of computer vision techniques, a large set of samples can be automatically measured, saving time and providing more objective and precise information. |
|---|