Answer Sheet Graphical Code Recognition for Chinese Characters

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing answer sheets are limited in functionality as they only support objective questions with English letters (A, B, C, D) and cannot recognize Chinese characters or subjective answers, restricting their application and preventing automatic scoring by computers.

Innovation Solution

A computer method that recognizes answer sheets using graphical codes, where each grid is divided into 16 triangular areas, with encoding format and function marks, allowing recognition of characters by converting binary codes into ASCII or Unicode characters based on encoding formats like GBK, UTF-8, and others.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the answer sheet uses traditional binary codes for objective questions, then the structure is simple and easy to process, but it cannot recognize Chinese characters or subjective answers, limiting its functionality

Engineering Contradiction:
Improvecharacter recognition capabilityVSAvoidcode structure complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The code structure is segmented into multiple functional components: a 4-bit region code for positioning, an 8-bit graphical code for character representation, and an 8-bit encoding format indicator. This segmentation allows the system to handle both traditional objective questions and Chinese characters/subjective answers by interpreting different bit patterns, thereby resolving the contradiction between versatility and complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The extended binary code structure serves multiple functions simultaneously: it can represent English letters for objective questions, Chinese characters for subjective answers, and encoding format information. By making the code structure universal, the system achieves both simplicity for traditional questions and expanded capability for modern requirements without creating separate systems.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Productivity

If the answer sheet supports only English letters, then the processing is simple and fast, but the applicable scope is limited and automatic scoring cannot be realized

Engineering Contradiction:
Improveautomatic marking efficiencyVSAvoidcharacter type support
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent replaces manual marking with an automated optical recognition system that reads the extended binary codes. The computer system automatically interprets the graphical codes and encoding format indicators to transcribe answers, enabling both English and Chinese character recognition, and subjective answer processing through the same automated mechanism, thus achieving high productivity with broad adaptability.

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

3Quantity of substance

If the answer sheet uses a compact code structure, then the information density is high, but the encoding format identification becomes complex

Engineering Contradiction:
Improveinformation densityVSAvoidencoding format detection
Core Design Contradiction:
Quantity of substanceVSDifficulty of detecting and measuring

Solution Approach 1:

The encoding format indicator is pre-positioned within the code structure itself, allowing the system to automatically identify the encoding format during the reading process. This preliminary inclusion of format information eliminates the need for separate detection steps, maintaining high information density while simplifying the detection process through integrated design.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11823001B2Computer implemented method for recognizing answer sheet
Publication Date: 2023.11.21 WANG YUHONG
  • US11823001B2 patent drawing
  • US11823001B2 patent drawing
  • US11823001B2 patent drawing

AI summary

A computer implemented method for recognizing an answer sheet includes the steps of: reading a filled answer sheet comprising at least one set of graphical codes; locating each set of graphical codes in the answer sheet; a set of graphical codes comprising one grid or two grids, wherein for each set of graphical codes, if the function mark of the first grid is filled, two sets or four sets of 8-digit binary codes are obtained according to the filling state of each of the filling areas in each grid of the set of graphical codes; and recognizing the corresponding characters according to the encoding format determined by the encoding format mark. If the function mark in the first grid is un-filled, the filling states of the eight icons of the encoding format mark are converted to the third set of binary codes for character recognition.