Abstract: Some embodiments of the invention include a method for updating an occlusion probability map. An occlusion probability map represents the probability that a given portion of the sensor field is occluded from one or more sensors. In some embodiments, a method may include receiving field of view data from a sensor system; producing a probabilistic model of the sensor field of view; and updating an occlusion probability map using the probabilistic model and field of view data.

Abstract: In one aspect, a system for illuminating a field of view of a vision-based sensor mounted on an agricultural machine may include an agricultural machine having a vision-based sensor. The system may also include a light source configured to emit supplemental light to illuminate at least a portion of the field of view of the vision-based sensor. Furthermore, the system may include a controller communicatively the light source. The controller may configured to control an operation of the light source based on an input indicative of ambient light present within the field of view of the vision-based sensor.

Abstract: The present disclosure provides systems and methods that measure crop residue in a field from imagery of the field. In particular, the present subject matter is directed to systems and methods that include or otherwise leverage a machine-learned crop residue classification model to determine a crop residue parameter value for a portion of a field based at least in part on imagery of such portion of the field captured by an imaging device. For example, the imaging device can be a camera positioned in a downward-facing direction and physically coupled to a work vehicle or an implement towed by the work vehicle through the field.

Abstract: Some embodiments of the invention include a method for updating an occlusion probability map. An occlusion probability map represents the probability that a given portion of the sensor field is occluded from one or more sensors. In some embodiments, a method may include receiving field of view data from a sensor system; producing a probabilistic model of the sensor field of view; and updating an occlusion probability map using the probabilistic model and field of view data.

Abstract: In one aspect, a system for adjusting a sampling rate of a sensor mounted on an agricultural machine may include an agricultural machine and a sensor mounted on the agricultural machine, with the sensor being configured to capture data at a sampling rate. The system may also include a controller communicatively coupled to the sensor. The controller may be configured to receive an input indicative of an operational parameter of the agricultural machine and adjust the sampling rate at which the sensor captures data based on the received input.

Abstract: The present disclosure provides systems and methods that measure crop residue in a field from imagery of the field. In particular, the present subject matter is directed to systems and methods that include or otherwise leverage a machine-learned crop residue classification model to determine a crop residue parameter value for a portion of a field based at least in part on imagery of such portion of the field captured by an imaging device. Furthermore, principal components analysis, such as projecting image patches onto Eigen-images, can be performed to reduce the dimensionality of the feature vector provided to the classification model.

Abstract: The present disclosure provides systems and methods that measure crop residue in a field from imagery of the field. In particular, the present subject matter is directed to systems and methods that include or otherwise leverage a machine-learned semantic segmentation model to determine a crop residue parameter value for a portion of a field based at least in part on imagery of such portion of the field captured by an imaging device. For example, the imaging device can be a camera positioned in a downward-facing direction and physically coupled to a work vehicle or an implement towed by the work vehicle through the field.

Abstract: The present disclosure provides systems and methods that measure crop residue in a field from imagery of the field. In particular, the present subject matter is directed to systems and methods that include or otherwise leverage a machine-learned convolutional neural network to determine a level of crop residue for a portion of a field based at least in part on imagery of such portion of the field captured by an imaging device. For example, the imaging device can be a camera positioned in a downward-facing direction and physically coupled to a work vehicle or an implement towed by the work vehicle through the field.

Abstract: The present disclosure provides systems and methods for adjusting the output of a field measurement system to conform to agronomy measurements. In particular, the present subject matter is directed to a calibration process and system that uses a calibration model to convert field measurement data expressed according to an automatic system metric into agronomy data that is expressed according to an agronomy metric.

Abstract: In one aspect, a system for monitoring sensor performance on an agricultural machine may include a controller configured to receive a plurality of images from the vision-based sensor mounted on an agricultural machine. The controller may be configured to determine an image parameter value associated with each of a plurality of pixels contained within each of the plurality of images. For each respective pixel of the plurality of pixels, the controller may be configured to determine a variance associated with the image parameter values for the respective pixel across the plurality of images. Furthermore, when the variance associated with the image parameter values for a given pixel of the plurality of pixels falls outside of a predetermined range, the controller may be configured to identify the given pixel as being at least one of obscured or inoperative.

Abstract: A method for automatically monitoring soil surface roughness as a ground-engaging operation is being performed within a field may include receiving pre-operation surface roughness data associated with a given portion of the field and receiving post-operation surface roughness data associated with the given portion of the field. In addition, the method may include analyzing the pre-operation and post-operation surface roughness data to determine a surface roughness differential associated with the performance of the ground-engaging operation and actively adjusting the operation of at least one of an associated work vehicle and/or implement when the surface roughness differential differs from a target set for the surface roughness differential.

Abstract: The present disclosure provides systems and methods that measure soil roughness in a field from imagery of the field. In particular, the present subject matter is directed to systems and methods that include or otherwise leverage a machine-learned clod detection model to determine a soil roughness value for a portion of a field based at least in part on imagery of such portion of the field captured by an imaging device. For example, the imaging device can be a camera positioned in a downward-facing direction and physically coupled to a work vehicle or an implement towed by the work vehicle through the field.

Abstract: The present disclosure provides systems and methods that measure soil roughness in a field from imagery of the field. In particular, the present subject matter is directed to systems and methods that include or otherwise leverage a machine-learned clod detection model to determine a soil roughness value for a portion of a field based at least in part on imagery of such portion of the field captured by an imaging device. For example, the imaging device can be a camera positioned in a downward-facing direction and physically coupled to a work vehicle or an implement towed by the work vehicle through the field.

Abstract: The present disclosure provides systems and methods that measure crop residue in a field from imagery of the field. In particular, the present subject matter is directed to systems and methods that include or otherwise leverage a machine-learned semantic segmentation model to determine a crop residue parameter value for a portion of a field based at least in part on imagery of such portion of the field captured by an imaging device. For example, the imaging device can be a camera positioned in a downward-facing direction and physically coupled to a work vehicle or an implement towed by the work vehicle through the field.

Abstract: In one aspect, a system for illuminating a field of view of a vision-based sensor mounted on an agricultural machine may include an agricultural machine having a vision-based sensor. The system may also include a light source configured to emit supplemental light to illuminate at least a portion of the field of view of the vision-based sensor. Furthermore, the system may include a controller communicatively the light source. The controller may configured to control an operation of the light source based on an input indicative of ambient light present within the field of view of the vision-based sensor.

Abstract: The present disclosure provides systems and methods for adjusting the output of a field measurement system to conform to agronomy measurements. In particular, the present subject matter is directed to a calibration process and system that uses a calibration model to convert field measurement data expressed according to an automatic system metric into agronomy data that is expressed according to an agronomy metric.

Abstract: The present disclosure provides systems and methods that measure crop residue in a field from imagery of the field. In particular, the present subject matter is directed to systems and methods that include or otherwise leverage a machine-learned crop residue classification model to determine a crop residue parameter value for a portion of a field based at least in part on imagery of such portion of the field captured by an imaging device. Furthermore, principal components analysis, such as projecting image patches onto Eigen-images, can be performed to reduce the dimensionality of the feature vector provided to the classification model.

Abstract: The present disclosure provides systems and methods that measure crop residue in a field from imagery of the field. In particular, the present subject matter is directed to systems and methods that include or otherwise leverage a machine-learned convolutional neural network to determine a level of crop residue for a portion of a field based at least in part on imagery of such portion of the field captured by an imaging device. For example, the imaging device can be a camera positioned in a downward-facing direction and physically coupled to a work vehicle or an implement towed by the work vehicle through the field.

Abstract: In one aspect, a system for adjusting a sampling rate of a sensor mounted on an agricultural machine may include an agricultural machine and a sensor mounted on the agricultural machine, with the sensor being configured to capture data at a sampling rate. The system may also include a controller communicatively coupled to the sensor. The controller may be configured to receive an input indicative of an operational parameter of the agricultural machine and adjust the sampling rate at which the sensor captures data based on the received input.

Abstract: In one aspect, a system for monitoring sensor performance on an agricultural machine may include a controller configured to receive a plurality of images from the vision-based sensor mounted on an agricultural machine. The controller may be configured to determine an image parameter value associated with each of a plurality of pixels contained within each of the plurality of images. For each respective pixel of the plurality of pixels, the controller may be configured to determine a variance associated with the image parameter values for the respective pixel across the plurality of images. Furthermore, when the variance associated with the image parameter values for a given pixel of the plurality of pixels falls outside of a predetermined range, the controller may be configured to identify the given pixel as being at least one of obscured or inoperative.