LiDAR360V9 - Classification | Classify Ground by Deep Learning

Hello everyone. This video introduces classify ground by deep learning function. Under the classification

function. Under the classification module, find the classify ground by deep learning function. After clicking this

learning function. After clicking this function, there are above parameters settings. The from class is the category

settings. The from class is the category to be classified which is the point to be classified. Default selection of all

be classified. Default selection of all your current categories. The two class is the category name of the classification result. Default is to

classification result. Default is to ground default. Click this button to restore

default. Click this button to restore all parameters to default. Use GPU first means choosing whether to use graphics card acceleration or not. This function

supports to running most GPU and CPU. If

the computer's graphics card meets the requirements, a video graphics card with compute capability of 3.5 or above and with more than 4 GB of available video

memory remaining during runtime. This

option will be checked by default. Users

can choose whether to use GPU as needed.

The GPU is about four times more efficient than the CPU. If the

conditions are not met, the GPU cannot be used. At this time, the function will

be used. At this time, the function will default to use CPU. After parameter

settings are completed, click okay to start classification.

After classification, select to display by classification. Only keep ground

by classification. Only keep ground points to view the classification of ground points. You can see that the

ground points. You can see that the classification effect is very good. The

above is the complete introduction to the function of classify ground by deep learning. Thanks for watching.

learning. Thanks for watching.