Image Division Unit for Document Chunking and Scene Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing information processing systems require reconstructing documents into structured information for search, which is costly and inefficient, leading to unnecessary information overload and slow processing.
Innovation Solution
An information processing device that acquires and divides images to estimate scenes and chunks using trained models, outputting relevant work information without the need for large-scale reconstruction, allowing for targeted and efficient information presentation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If documents are reconstructed into structured information for search, then search accuracy is improved, but processing cost and time increase significantly
Solution Approach 1:
The patent segments documents into smaller chunks based on image boundaries and text regions, allowing partial processing rather than complete document reconstruction. This enables search on specific relevant portions while avoiding processing entire documents, thus improving search accuracy for target information while reducing processing time and costs.
Solution Approach 2:
The patent applies partial action by processing only the necessary portions of documents (specific chunks identified through image analysis) rather than reconstructing entire documents. This selective processing approach maintains search accuracy for relevant information while significantly reducing the time and resources required compared to complete document reconstruction.
2Loss of information
If all documents are reconstructed into structured information, then information completeness is improved, but processing cost becomes unrealistic
Solution Approach 1:
The patent divides documents into multiple chunks based on image boundaries and text regions, allowing the system to process and store only essential portions. This segmentation enables maintaining information completeness for search-relevant content while avoiding the prohibitive costs of reconstructing entire documents, making the approach economically viable.
Solution Approach 2:
The patent extracts and processes only the necessary information chunks from documents based on image analysis results, rather than reconstructing complete documents. This extraction approach ensures that essential information is preserved and searchable while significantly reducing processing costs to realistic levels.
3Loss of information
If information is provided only in document units, then information completeness is improved, but information overload increases and handling speed decreases
Solution Approach 1:
The patent segments documents into smaller, manageable chunks based on image boundaries and text regions. This allows the system to present only relevant information portions to users rather than entire documents, reducing information overload while maintaining completeness of essential information, and enabling faster handling and retrieval.
Solution Approach 2:
The patent extracts and presents only the necessary information chunks relevant to user queries, rather than providing complete documents. This extraction approach reduces information overload for users, improves handling speed by eliminating unnecessary content, while still maintaining information completeness for the essential portions needed.
Data Source
AI summary
An information processing device is configured to output work information related to work performed by a serving person, the information processing device including an image acquisition unit configured to acquire an original image including a served person and a plurality of served objects that the serving person serves, an image division unit configured to divide the original image into a served-person image, in which the served person is captured, and a plurality of served-object images, in which each served object is captured, a scene estimation unit configured to estimate a scene, which is the situation the serving person is in, by using a first trained model, a chunk estimation unit configured to estimate a chunk, which is information dividing or suggesting the work information, by using one of a plurality of second trained models, and an output unit configured to output the chunk.