ChartFuse: a novel fusion method for chart classification using heterogeneous microstructures
- PDF / 2,263,451 Bytes
- 23 Pages / 439.37 x 666.142 pts Page_size
- 13 Downloads / 192 Views
ChartFuse: a novel fusion method for chart classification using heterogeneous microstructures Prerna Mishra 1 & Santosh Kumar 1 & Mithilesh Kumar Chaube 2 Received: 7 May 2020 / Revised: 3 September 2020 / Accepted: 11 November 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
Chart images exhibit significant structural variabilities, which is called as micro-structural variabilities, which makes each image type different from others even though chart image belongs to the same class or categories. The lack of affiliation between the heterogeneous features and the structure of the chart images, make it challenging to learn these microvariabilities features by any learning model for automatic chart recognition and interpretation. However, extracting low-level heterogeneous features from chart images remains challenging. This paper presents a novel chart image classification method by using local feature descriptor. We proposed a new heterogeneous feature extractor, namely the heterogeneity index (HI) fused with local penta pattern. Here, the microstructural features are defined on the similarity of the chroma effects, and HI is computed depending upon the basic colors and its intensity in the microstructures with similar chroma effects. HI integrates colors, textures, structural layout, and illumination details with the local features altogether for the image classification. The proposed method is tested on chart image datasets, namely FigureQA and our handcrafted chart dataset. Experimental results depict that our method classify images with an accuracy rate of 95%–97% which is an increase of 5%–10% as compared with the customary methods. Keywords Chart interpretation . Chart recognition . Microstructural features . Heterogeneous local penta pattern . Heterogeneous index
* Prerna Mishra [email protected] Santosh Kumar [email protected] Mithilesh Kumar Chaube [email protected]
1
Department of Computer Science and Engineering, DSPM-IIITNR, Raipur, India
2
Deparment of Mathematics, DSPM-IIITNR, Raipur, India
Multimedia Tools and Applications
1 Introduction Charts are the easiest mechanism to represent and visualize data [10, 24]. Due to the variation in data and its classes, data can be represented in various forms. Each chart type consists of both heterogeneous and homogeneous features of different classes. This representation of chart data varies from class-to-class. These chart variations are majorly due to the orientation of charts, chart data, different chart styles and appearances [41, 56]. Extraction and interpretation of chart data is an important step in any learning algorithm. Due to these variations, various interpretational ways can entangle and hide more or less different explanatory factors of variations behind the chart data. For data interpretation, feature extraction and learning are necessary steps [26, 38]. However, current learning algorithms are not properly utilized to transform discriminatory features of charts. Moreover, existing feature ex
Data Loading...
