Conventional image categorization techniques primarily rely on low-level visual cues. In this paper, we describe a multimodal fusion scheme which improves the image classification...
A central problem in information retrieval is the automated classification of text documents. While many existing methods achieve good levels of performance, they generally require...
This paper describes a method of readability measurement of Japanese texts based on a newly compiled textbook corpus. The textbook corpus consists of 1,478 sample passages extract...
Automatic text categorization is a problem of automatically assigning text documents to predefined categories. In order to classify text documents, we must extract good features f...
In this paper, we propose a new method based on wavelet transform, statistical features and central moments for both graphics and scene text detection in video images. The method ...
Palaiahnakote Shivakumara, Trung Quy Phan, Chew Li...