In the proposed system, detected data from the wearable device is analyzed at a remote site instead of by an internal computer. By separating this function, the wearable device becomes a simple, small, and lightweight unit with high performance. By using the detected pattern and sensing data, farming operations are recognized with estimation algorithms such as pattern matching, Bayesian estimation, principal component analysis, and support vector machines which classify the data in groups of
farming operation with supervised learning. This recognition function is easily and effectively modified because of the distributed processing architecture. This distributed processing architecture
also enables interactive support applications. By using Field Servers and peripheral devices, these applications control suitable machines in a coordinated manner, monitor various conditions in detail,
and provide useful information to a farmer in response to recognized farming operations.
In the proposed system, detected data from the wearable device is analyzed at a remote site instead of by an internal computer. By separating this function, the wearable device becomes a simple, small, and lightweight unit with high performance. By using the detected pattern and sensing data, farming operations are recognized with estimation algorithms such as pattern matching, Bayesian estimation, principal component analysis, and support vector machines which classify the data in groups of farming operation with supervised learning. This recognition function is easily and effectively modified because of the distributed processing architecture. This distributed processing architecture also enables interactive support applications. By using Field Servers and peripheral devices, these applications control suitable machines in a coordinated manner, monitor various conditions in detail, and provide useful information to a farmer in response to recognized farming operations.
การแปล กรุณารอสักครู่..