Document creation support apparatus, document creation support method, and program
AssigneeFUJIFILM Corporation
InventorKeigo NAKAMURA
Life of this patent
Estimating the term…
Abstract
A first text generation unit 15 generates a first text describing properties of an abnormal shadow for each of abnormal shadows included in a diagnosis target image. A classification unit 14 classifies a plurality of abnormal shadows into at least one group. A second text generation unit 16 generates a second text integrating the first text generated for each of the abnormal shadows belonging to the same group, for each group. A third text generation unit 17 generates a third text integrating the second text generated for each group.
Claims12
A document creation support apparatus comprising at least one processor,
- wherein the processor is configured to
- generate a first text describing a property of a feature portion for each of a plurality of feature portions included in an image,
- classify the plurality of feature portions into at least one group,
- generate a second text integrating the first text generated for each of the feature portions belonging to the same group for each group, and
- generate a third text integrating the second text generated for each group.
A document creation support method comprising:
- generating a first text describing a property of a feature portion for each of a plurality of feature portions included in an image;
- classifying the plurality of feature portions into at least one group;
- generating a second text integrating the first text generated for each of the feature portions belonging to the same group for each group;
- and
- generating a third text integrating the second text generated for each group.
A non-transitory computer-readable storage medium storing a program for causing a computer to execute a process comprising:
- generating a first text describing a property of a feature portion for each of a plurality of feature portions included in an image;
- classifying the plurality of feature portions into at least one group;
- generating a second text integrating the first text generated for each of the feature portions belonging to the same group for each group;
- and
- generating a third text integrating the second text generated for each group.
Dependencies are inferred from the claim text.