Extracting Feature Values from Point Clouds to Generate Plant Treatments

A non-transitory computer readable storage medium comprising instructions that, when executed by the processor, cause the processor to:

• access a point cloud comprising a plurality of points, each of the points labelled as a plant point or a ground point and having a three-dimensional coordinate such that the point cloud represents a three-dimensional scene in the field;
• apply one or more pre-processing functions to the point cloud, the pre-processing functions:
• determining one or more characteristics for the three-dimensional scene in the field, and
• modifying a label or a three-dimensional coordinate of one or more points in the point cloud;
• identify a cluster of points in the point cloud representing one or more plants based on the labels and the three-dimensional coordinates of the cluster of points;
• derive a feature value for the one or more plants based on the three-dimensional coordinates of the points in the cluster and the one or more extracted characteristics for the three-dimensional scene in the field;
• and
• in response to the derived feature value indicating at least one plant of the one or more plants should be treated, actuate a treatment mechanism of the plurality of treatment mechanisms to treat the at least one plant as the farming machine travels past at least one plant in the field.

A farming machine for treating a plant in the field as the farming machine moves through the field, the farming machine comprising:

• a plurality of treatment mechanisms to treat the plant as the farming machine travels past the plant in the field;
• a processor;
• and
• a non-transitory computer readable storage medium comprising instructions that, when executed by the processor, cause the processor to:
• access a point cloud comprising a plurality of points, each of the points labelled as a plant point or a ground point and having a three-dimensional coordinate such that the point cloud represents a three-dimensional scene in the field;
• apply one or more pre-processing functions to the point cloud, the pre-processing functions:
• determining one or more characteristics for the three-dimensional scene in the field, and
• modifying a label or a three-dimensional coordinate of one or more points in the point cloud;
• identify a cluster of points in the point cloud representing one or more plants based on the labels and the three-dimensional coordinates of the cluster of points;
• derive a feature value for the one or more plants based on the three-dimensional coordinates of the points in the cluster and the one or more extracted characteristics for the three-dimensional scene in the field;
• and
• in response to the derived feature value indicating at least one plant of the one or more plants should be treated, actuate a treatment mechanism of the plurality of treatment mechanisms to treat the at least one plant as the farming machine travels past at least one plant in the field.