Publikace: Simplified version of white wine grape berries detector based on SVM and HOG features
Konferenční objektOmezený přístuppeer-reviewedpostprintNačítá se...
Soubory
Datum
Autoři
Škrabánek Pavel
Majerík Filip
Název časopisu
ISSN časopisu
Název svazku
Nakladatel
Springer International Publishing AG
Abstrakt
The detection of grapes in real scene images is a serious task solved by researches dealing with precision viticulture. Our research has shown that in the case of white wine varieties, grape berry detectors based on a support vector machine classifier in combination with a HOG descriptor are very efficient. In this paper, simplified versions of our original solutions are introduced. Our research showed that skipping contrast normalization by image preprocessing accelerates the detection process; however, the performance of the detectors is not negatively influenced by this modification.
Popis
Klíčová slova
computer vision, precision viticulture, grape detection, HOG features, support vector machine, počítačové vidění, precisní vinohradnictví, detekce hroznů, HOG deskriptor, support vector machine