Abstract: A method of board lumber grading is performed in an industrial environment on a machine learning framework configured as an interface to a machine learning-based deep convolutional network that is trained end-to-end, pixels-to-pixels on semantic segmentation. The method uses deep learning techniques that are applied to semantic segmentation to delineate board lumber characteristics, including their sizes and boundaries.

Abstract: A method of board lumber grading is performed in an industrial environment on a machine learning framework configured as an interface to a machine learning-based deep convolutional network that is trained end-to-end, pixels-to-pixels on semantic segmentation. The method uses deep learning techniques that are applied to semantic segmentation to delineate board lumber characteristics, including their sizes and boundaries.

Abstract: A method of board lumber grading is performed in an industrial environment on a machine learning framework configured as an interface to a machine learning-based deep convolutional network that is trained end-to-end, pixels-to-pixels on semantic segmentation. The method uses deep learning techniques that are applied to semantic segmentation to delineate board lumber characteristics, including their sizes and boundaries.

Abstract: A method of improving board feature quality grading facilitates human check grader reaction to grading output. The method entails specifying, for application by an automatic grading system to the faces of boards passing through a scanning zone, a virtual grade expressed by a range of values that overlap values in two of successive standard rule-based grades representing higher and lower board feature qualities; producing a signal in response to detection of a board feature quality representing a virtual grade value of a board analyzed by the automatic grading system; indicating, in response to the signal, a virtual grading designation onto the board to alert a check grader to assess whether the board exhibits board feature quality that exceeds the lower board feature quality; and assigning to the board the standard rule-based grade representing the higher board feature quality whenever the check grader's assessment concurs with the virtual grade designation.

Abstract: A method of board lumber (Table 2) grading is performed in an industrial environment on a machine learning framework (12) configured as an interface to a machine learning-based deep convolutional network (20) that is trained end-to-end, pixels-to-pixels on semantic segmentation. The method uses deep learning techniques that are applied to semantic segmentation to delineate board lumber characteristics (Table 1), including their sizes and boundaries.

Abstract: A method of improving board feature quality grading facilitates human check grader reaction to grading output. The method entails specifying, for application by an automatic grading system to the faces of boards passing through a scanning zone, a virtual grade expressed by a range of values that overlap values in two of successive standard rule-based grades representing higher and lower board feature qualities; producing a signal in response to detection of a board feature quality representing a virtual grade value of a board analyzed by the automatic grading system; indicating, in response to the signal, a virtual grading designation onto the board to alert a check grader to assess whether the board exhibits board feature quality that exceeds the lower board feature quality; and assigning to the board the standard rule-based grade representing the higher board feature quality whenever the check grader's assessment concurs with the virtual grade designation.