Scribd
Upload
8 views4 pages

Automated Graph Reader

The Automated Graph Reader (AGR) project focuses on developing a web or desktop application to extract data from graphical images for data mining purposes. The project is divided into two phases: a prototype phase that implements basic functionalities and a final application phase that enhances automation and batch processing capabilities. Key features include image processing, data point selection, and data persistence using various programming languages and tools.

Uploaded by

Sunwar Khan 813-FBAS/MSCS/F14
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
8 views4 pages

Automated Graph Reader

The Automated Graph Reader (AGR) project focuses on developing a web or desktop application to extract data from graphical images for data mining purposes. The project is divided into two phases: a prototype phase that implements basic functionalities and a final application phase that enhances automation and batch processing capabilities. Key features include image processing, data point selection, and data persistence using various programming languages and tools.

Uploaded by

Sunwar Khan 813-FBAS/MSCS/F14
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 4

Automated Graph Reader (AGR)

Project Domain / Category


Software Engineering/Web/Desktop/Machine learning/Research.

Abstract / Introduction
Data mining has several real-world applications, and data acquisition is the first step in the process
of analysis and perform decision-making tasks. Sometimes, data is only available in graphical
images that may require costly computational methods to process. However, the computational
complexity can be reduced by extracting the data from graph images and then using numeric data
instead of images in future phases of analysis and prediction. In this project, students are required
to design and develop a Web or Desktop Application in the language of their choice to extract
graph data from images.

Figure 1 - Graph Reading

Figure 2 - Graph Components


Functional Requirements:
The functional requirements are grouped into two phases. In the first phase, students will develop
a prototype following specific requirements. The requirements in the first phases are
implementation of basic functionality. In second phase, students will develop the final version of
the web or desktop application with additional requirements as specified below. The requirements
of the second phase mostly focus on automatic detection and extraction of data and batch
processing.
1.1 PROTOTYPE PHASE
1.1.1 Image Processing
The solution should allow processing the input images in terms of cropping, changing colors,
and image improvement.

1.1.2 Image Cropping and Alignment


 Crop additional white space around graphs' sides manually in batch mode.
 Align the graph's sides manually in batch mode.
 Align the graph's sides manually in batch mode automatically.

1.1.3 Image Scaling and axis detection


 Allow users to manually specify the scale of the graph by selecting the corner points
and assigning the values.

1.1.4 Data Point Selection

 Allow manual selection of data points and calculate the coordinates automatically.
 Data shapes
o Circle Filled (●)
o Asterisk (*)
o Cross (x)
 Graph Types
o The solution should be able to work on the following graphs at least.
Line Bar Network

Figure 3 - Graph Types


1.1.5 Data Persistence
 Store all extracted data against the user's specified data.
 All the operations must be interactive.
1.1.6 Processing Flow
 Upload Images
 Preprocess
 Axis selection and Scaling
 Datapoint Selection/Detection
 Save Data
 Process next image

1.2 FINAL APPLICATION


1.2.1 Image Cropping and Alignment
 Crop additional white space around graph sides automatically

1.2.2 Image Scaling and axis detection


 Allow automated scale rendering by reading the values of the image graph. No user
input should be required for this feature.

1.2.3 Data Point Selection


 Data shapes
o All distinct marker shapes.
 Graph Types
o The solution should be able to work on all graph types.
 Detect the graph artifacts, suggest suitable data points, and calculate
their coordinates automatically.

1.2.4 Data Persistence


o Persist all extracted data in MySQL/Sqlite/MongoDB.

1.2.5 Batch Processing


 Batch Uploading Images
 Batch Preprocessing
 Automated Axis selection and Scaling
 Automated Datapoint Selection/Detection

Tools:
Languages:
Python, PHP,
C/C++ IDE: IDE of
choice
OS: Linux/Window of Choice
Graph Repository: any, Example :
https://github.com/JasonObeid/Chart2TextImages

You might also like