Chart Data Digitization via Automated Element Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Scientific charts in published papers often present unstructured data, making it difficult for readers to extract numeric values, convert between measurement systems, or replicate charts with different styles and scales, as manual measurement methods are cumbersome and lack automation.
Innovation Solution
A system for real-time scientific chart classification and digitization (CCD) that allows users to select and copy chart data, automatically identifying and converting chart elements into numeric values and generating data tables, enabling easy pasting into documents and conversion between chart types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If manual measurement methods are used to extract data from charts, then data extraction is possible, but the process is cumbersome and lacks automation
Solution Approach 1:
The system enables self-service by allowing users to simply select and copy chart data, with the automated processing occurring in the background. The CCD manager automatically identifies chart elements, performs measurements, and converts data to structured formats without requiring manual intervention for each step.
Solution Approach 2:
The patent replaces manual mechanical measurement operations with an automated computer vision and image processing system. The system uses optical character recognition, edge detection, and coordinate transformation algorithms to automatically extract numeric values from chart images, substituting the need for physical rulers and manual measurement.
2Loss of information
If chart data is presented in visual chart format, then data is space-effective and easy to understand, but numeric values are difficult to extract
Solution Approach 1:
The system extracts numeric values from the visual chart representation and separates them into a structured data table format. The CCD manager identifies chart elements such as axes, labels, and data points, then extracts the numeric information associated with these elements, making the data accessible for further processing while preserving the original visual chart for reference.
Solution Approach 2:
The system changes the parameter representation of chart data from visual/spatial parameters (positions, lengths on the chart image) to numeric data parameters (actual measurement values, scaled data). The coordinate transformation process converts image coordinates to chart data coordinates, enabling the extraction of meaningful numeric values from visual representations.
3Adaptability or versatility
If conversion between measurement systems is needed, then data versatility is improved, but manual conversion is time-consuming
Solution Approach 1:
The system performs preliminary action by capturing the measurement system and scale information from the chart during the initial data extraction phase. The CCD manager identifies axis labels, units, and scale factors, storing this metadata alongside the extracted numeric values. This preliminary capture of conversion information enables rapid unit conversions later without requiring manual intervention.
4Adaptability or versatility
If chart replication with different styles and scales is required, then chart adaptability is improved, but manual recreation is cumbersome
Solution Approach 1:
The system enables easy copying by providing the extracted numeric data and chart parameters in structured formats that can be readily used to recreate charts. The data table contains all necessary information including numeric values, axis labels, scale factors, and chart type information, which can be used to generate new charts with different styles and scales without manual recreation efforts.
Data Source
AI summary
Embodiments are disclosed for generating tables from charts. The techniques include determining a chart type of a selected chart. The techniques include determining a plurality of chart elements of the selected chart. Further, the techniques include determining a plurality of measurements of a plurality of data representations of the selected chart. Additionally, the techniques include determining a plurality of corresponding numeric values for the measurements based on the chart elements. Also, the techniques include generating a data table comprising the chart elements and the corresponding numeric values.