Indra Kumari and Hansung Lee ,  IEEE ACCESS , vol. 13, 101789 - 101800

Abstract: Chart-to-text conversion is an emerging research area focused on extracting useful information from chart images to improve understanding and analysis. Deep learning methods help in identifying important details and patterns from charts. However, existing models struggle to analyze charts because of the mix of text and graphical elements. To solve this problem, we propose a new method called MultiSHTM (Multi-level Stacked Houghless Network-based Bi-LSTM). This method improves accuracy, reduces complexity, and works well across different chart types. MultiSHTM integrates two key innovations: 1) a multilevel attention mechanism in a stacked houghless network, which accurately identifies key points in charts without relying on traditional Hough Transform-based methods and 2) a Bi-LSTM model enhanced with a Hierarchical and Channel Attention module, which effectively captures contextual relationships to generate precise summaries of chart images. Compared to existing methods, MultiSHTM performs better, achieving scores of Rouge: 0.55, Bleu: 0.45, Cider: 0.8, Meteor: 0.25, and Spice: 25.60.

Min-Seok Kim, Ye-Ju Kim, Chan-Hyeok Lee, Hansung Lee ,  in Proceedings of Korea Computer Education Conference , vol. 28, no. 2.

Abstract(Short): Grading and evaluation of apples is a crucial process in agricultural quality control and an important step in determining the price of apples. The visual inspection for apple grading and evaluation assesses the size, shape, and color of apples to determine their grade, and currently relies entirely on manual inspection. As a result, consistency in grading and evaluation can be difficult to achieve due to the inspector's level of experience and fatigue. This study aims to address these issues and seeks to develop an automated inspection method to perform consistent visual assessments. The study explores the feasibility of automating apple grading based on visual inspection using machine learning techniques. The possibility of grade determination based on visual inspection using machine learning was verified using apple grade data collected from farms in the Gyeongbuk region.