Key Frame Extraction for Text Based Video Retrieval Using Maximally Stable Extremal Regions
Journal Title: EAI Endorsed Transactions on e-Learning - Year 2015, Vol 2, Issue 7
Abstract
This paper presents a new approach for text-based video content retrieval system. The proposed scheme consists of three main processes that are key frame extraction, text localization and keyword matching. For the key-frame extraction, we proposed a Maximally Stable Extremal Region (MSER) based feature which is oriented to segment shots of the video with different text contents. In text localization process, in order to form the text lines, the MSERs in each key frame are clustered based on their similarity in position, size, color, and stroke width. Then, Tesseract OCR engine is used for recognizing the text regions. In this work, to improve the recognition results, we input four images obtained from different pre-processing methods to Tesseract engine. Finally, the target keyword for querying is matched with OCR results based on an approximate string search scheme. The experiment shows that, by using the MSER feature, the videos can be segmented by using efficient number of shots and provide the better precision and recall in comparison with a sum of absolute difference and edge based method.
Authors and Affiliations
Werachard Wattanarachothai, Karn Patanukhom
Keywords
Related Articles
- EP ID EP45951
- DOI http://dx.doi.org/10.4108/icst.iniscom.2015.258410
- Views 290
- Downloads 0
Collaboration on the web – Chances of participation in a formal education context
This contribution addresses the options and chances for an alliance between students’ participation and the adoption of digital media in a context of academic education at universities. Under certain circumstances these...
A Method for Teaching the Modeling of Manikins Suitable for Third-Person 3-D Virtual Worlds and Games
Virtual Worlds have the potential to transform the way people learn, work, and play. With the emerging fields of service science and design science, professors and students at universities are in a unique position to lea...
Blended learning for developing effective virtual teams: a proposed intervention format
The aim of this exploratory study was to develop a blended learning approach to fostering the skills and competencies required by leaders and members of international virtual teams. Three levels of analysis were brought...
The use of digital educational resources in the support to learning in higher education
This paper aimed to assess the importance given to the use of digital educational resources, their use frequency, and their classification considering them as a support to course units. The data was obtained through ques...
Keep me posted! Human and machine learning analysis of Facebook updates
The key element of Facebook social network platform is the status updates, in which the user can upload text or other media such as pictures and videos. In this study, we manually classified more than 3500 Facebook statu...