Automated Examination Paper Scoring via Image Region Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current systems for recognizing entry remarks on examination papers require manual specification of character recognition locations, increasing the workload for users such as teaching staff, as they need to pre-define the location for scoring each question, which is inefficient.

Innovation Solution

An information processing apparatus and method that automatically detects the regions of question-answer pairs and partial scores on examination papers using image processing, without pre-specifying the recognition locations, allowing for automated association and aggregation of scores.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the location of character recognition is specified in advance for each question, then the character recognition can be performed at the specified location to determine the score of each question, but the workload of the user increases

Engineering Contradiction:
Improvecharacter recognition accuracyVSAvoiduser workload
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The system performs self-service by automatically detecting the location of entry remarks on the original document without requiring user specification. The character recognition unit autonomously identifies and processes scores at their actual positions on examination papers, eliminating the need for users to pre-define recognition locations for each question.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If the location of character recognition is specified on a per entry item basis, then the score of each question can be determined accurately, but the time required for processing increases

Engineering Contradiction:
Improvescore determination accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary action by automatically detecting and storing the locations of entry remarks before character recognition is performed. The position information is extracted in advance, allowing the character recognition unit to efficiently process scores without time-consuming manual location specification for each question.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If manual specification of recognition locations is required, then character recognition can be performed precisely, but the complexity of the operation increases

Engineering Contradiction:
Improverecognition location precisionVSAvoidoperation complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system replaces the mechanical process of manual location specification with an automated image processing and character recognition system. The position information extraction unit automatically detects entry remark locations using image processing techniques, substituting manual operational complexity with automated computational processes that maintain precision without increasing operational burden.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP3142041B1Information processing apparatus, information processing method and program
Publication Date: 2021.08.11 FUJIFILM BUSINESS INNOVATION CORP
  • EP3142041B1 patent drawingFigure 1
  • EP3142041B1 patent drawingFigure 2
  • EP3142041B1 patent drawingFigure 3

AI summary

An information processing apparatus includes an acquisition unit that acquires image information of an image of an original document, a recognition unit that recognizes from the image information acquired by the acquisition unit each of plural entry items and an entry remark written differently from the plural entry items, included in the image, and an associating unit that associates the entry remark with one of the entry items in accordance with a relationship between a location of the entry items recognized by the recognition unit and a location of the entry remark.