A Convolutional Neural Network Network-Based Based Chinese Text Detection Algorithm via Text Structure Modeling
Abstract: Text detection in a natural environment plays an important role in many computer vision applications. While existing text detection methods are focused on English characters, there are strong application demands on text detection in other languages, such ass Chinese. In this paper, we present a novel text detection algorithm for Chinese characters based on a specific designed convolutional neural network (CNN). The CNN contains a text structure component detector layer, a spatial pyramid layer, and a multi multi-input-layer layer deep belief network (DBN). The CNN is pre-trained trained via a convolutional sparse auto auto-encoder, encoder, specifically designed for extracting complex features from Chinese characters. In particular, the text structure component detectors enhance the accuracy and uniqueness of feature descriptors by extracting multiple text structure components in various ways. The spatial pyramid layer enhances the scale invariability of the CNN for detecting texts in multiple scales. Finally, the multi multi-input-layer layer DBN replaces replace the fully connected layers in the CNN to ensure features from multiple scales are comparable. A multilingual text detection dataset, in which texts in Chinese, English, and digits are labeled separately, is set up to evaluate the proposed text detection algorithm. The proposed algorithm shows a significant performance improvement over the baseline CNN algorithms. In addition the proposed algorithm is evaluated over a public multilingual benchmark and achieves statestate