Detection of oil rape seed losses before harvest by Image analysis within Fog computing

Losses in the production of oilseed rape can occur before harvest, caused by spontaneous opening mature pods and fall seeds on ground. Different seed pods, among other things, ripen same plant at different times. So, cultivation similar crops, one most critical moments is determining right time for harvesting, because late harvest implies overripeness shell, which leads to spoilage, losses mechanical damage during threshing. One way monitoring shedding thus potential placing a container between rows plants monitor number that from open shells. The presented model system, consists sensor devices with associated cameras, positioned above position, has function transmit images current state. Central this paper an image analysis application be performed near sites computer-aided within Fog Computing. In way, results are obtained almost immediately forwarded Cloud platform or directly user who will take appropriate action. By obtaining timely information scattered seeds, it possible organize optimal order avoid prevent over-ripeness rape.