System and method for generating a sortation module and carcass processing options
AssigneeTriumph Foods, LLC
InventorsEmily Arkfeld, Eric Psota, Kenneth Grannas, III, Matthew England, Barry Wiseman, Stanley Stohr
Life of this patent
Estimating the term…
Abstract
The disclosure relates to methods for generating carcass processing options based on a primal/subprimal weights and yield training module. In one embodiment, the methods disclosed herein use automated methods to process image data using one or more trained modules, and determine carcass processing options based on an estimated weight calculation output from the trained modules, without using a scale-based weight measurement as an input.
Claims20
A method for characterizing carcass processing options, the process comprising:
- compiling an input data set;
- compiling a training data set, the training data set having one of a plurality of data types and a known classification value associated with the one of the plurality of data types;
- training one of a plurality of raw training modules with the training data set by iteratively:
- inputting the training data sets into the one of the plurality of raw training modules based on the one of the plurality of data types;
- comparing outputs of the one of the plurality of raw training modules to the known classification value for the training data set input into the one of the plurality of raw training modules;
- and
- updating one or more classification guidelines for a plurality of nodes of the one of the plurality of raw training modules based on results of the comparing step;
- and
- determining the carcass processing options based on an estimated weight output from at least one of a trained module of the plurality of raw training modules;
- and
- transmitting the carcass processing options to a processing unit.
A method for characterizing carcass processing options, the process comprising:
- compiling an input data set comprising scan image data;
- compiling a training data set including a plurality of data types and a classification value associated with each of the plurality of data types;
- iteratively training a plurality of raw training modules with the training data set;
- generating a trained module;
- processing the input data set into the trained module;
- determining the carcass processing options based on an estimated weight calculation output from the trained module;
- and
- transmitting the carcass processing options to a processing unit.
A method for characterizing carcass processing options, the process comprising:
- compiling an input data set comprising scan image data;
- compiling a training data set including a plurality of data types and a classification value associated with each of the plurality of data types;
- generating a trained module using the training data set;
- processing the input data set into the trained module;
- determining the carcass processing options based on an estimated weight calculation output from the trained module, wherein the estimated weight calculation output is calculated without the use of a scale-based weight measurement input;
- and
- transmitting the carcass processing options to a processing unit.
Dependencies are inferred from the claim text.