Chart Descriptor
In multiple fields charts have been used to illustrate data. In the actual world, charts are used to depict entities and their relationships in a range of domains, such as social networks, e-commerce networks, and physics networks. Charts are also noted for their complex layouts, which can hold huge
2025-06-28 16:25:48 - Adil Khan
Chart Descriptor
Project Area of Specialization Computer ScienceProject SummaryIn multiple fields charts have been used to illustrate data. In the actual world, charts are used to depict entities and their relationships in a range of domains, such as social networks, e-commerce networks, and physics networks. Charts are also noted for their complex layouts, which can hold huge underlying values. Data visualization through charts is much easier to understand than text. From large-scale data reading and analyzing the chart is a very hectic task to perform. Charts are mostly used to gather information and report, people face many problems in writing descriptions for charts, because when we have a large number of charts and data sets then there are more chances of errors, which leads to a bad representation of charts. We need such kind of system which can automatically generate descriptions for a certain chart. Charts are very supportive in understanding a huge sum of information or data by showing the findings in a chart. However, these graphs can get complex and need a description that explains in a manner that can be easily understood. The previous practices of manually writing graph descriptions do not give accurate results because there are more chances of errors, along with this issue the main problem was that a small number of charts can be described easily through the manual process but it is very difficult to manually describe a large number of charts and also the process is very time taking, so these issues have prompted a need to develop a solution that can automatically create graph descriptions with the help of an algorithm on a web-based application. Chart Descriptor will train a model which will classify charts from a certain image and then it will tell the type of the chart. Chart descriptor read the chart and converts it into text with the help of the Natural language processing technique (NLP) and it will provide a platform to automatically generate descriptions of charts rather than writing the text manually. People regularly use charts to help them grasp enormous volumes of data and the relationships between the data’s various components. Charts are frequently easier to understand than raw data. They’re used in a wide range of fields like medical diagnoses, weather forecasting, industries, education system, etc. The Chart Descriptor presents a new framework for automatically creating charts and analytical descriptions. The system simply requires an image of the chart, after which it will automatically generate a description. By using this system there will be fewer errors in graph findings and descriptions. People working in a variety of fields will have their workload decreased as a result of this. Chart descriptor will also provide a feature of audio of the description so that if a person is blind then he can easily hear the audio of the description. With the help of the Chart Descriptor system, people can easily get their graph descriptions automatically and can save time.
Project Objectives” A single plate-form for multiple chart descriptions”. Our goal is to provide a system that will generate descriptions from given chart images this will increase the efficiency and integrity of the chart content. It will also make digital content more accessible to persons with visual problems.
The objectives of this study are as follows:
- The aim of this system, Chart Descriptor is to train thr model to classify charts from images, and then it will decide the type of the chart.
- The aim is to make this chart descriptor to read the chart and convert it into text with the help of the Natural language processing technique (NLP).
Another aim of Chart Descriptor is to provide a platform to automatically generate descriptions of charts rather than writing the text manually.
Project Implementation MethodDue to the massive integration of information technology in all areas of modern life, people are wishing for more technologies and systems which can make their work easier, fastly, and more precisely. People working in large-scale industries and other sectors have to go through a large amount of data which contains numerous charts having deep descriptions, but most the people still face many problems in reading and understanding graphs although manually describing charts is very difficult to work.
Firstly, the user will provide an image, and the chart descriptor will perform machine learning to classify the chart type. To read the image of the chart we will apply CNN algorithm and OCR techniques. To classify between charts, we will consider
- Chart title
- X-axis
- Y-axis
- Y-axis
If the image has multi-panel charts, then the system will also be able to classify between the multi-panel charts. After the classification process now, the system will read the chart and convert it into text with the help of the Natural language processing technique (NLP). The system will generate an Analytical description of the charts by reading the chart data.
Benefits of the Project- Users will get their chart descriptions within seconds by just uploading a chart picture.
- Chart Descriptor will provide a platform to easily classify between charts.
- Chart Descriptor will benefit all the people worldwide who will be working on chart descriptions.
- Chart Descriptor will save time rather than manually writing descriptions for charts.
- Chart Descriptor will help those people who have visualizing problems, they will easily understand the chart by just reading the description.
- Chart Descriptor will generate accurate and efficient descriptions as compared to humans.
The final deliverable of "Chart Descriptor" will be a web-based application along with a documentation manual having all the technical details of the application.
Final Deliverable of the Project Software SystemCore Industry ITOther Industries Education , Medical , Petroleum , Manufacturing , Media , Telecommunication Core Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Quality Education, Decent Work and Economic Growth, Industry, Innovation and InfrastructureRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 75000 | |||
| PC with 1TB Hard Drive | Equipment | 1 | 70000 | 70000 |
| Printing | Miscellaneous | 1 | 5000 | 5000 |