Video Artificial Text Detection And Extraction

With the rapid development of the internet,communication technology,and smart phones,video acquisition and transmission become very convenient.Meanwhile,the video itself has the advantages of abundant information and comfort.Various factors make video replace traditional media as the most popular media.Now,the number of online videos has increased dramatically because of the convenience of uploading and downloading videos.Accordingly,there is a high demand for efficient indexing,retrieval,and localization of desired content from massive videos.Video text can depict the video content more directly and accurately compared with low-level perceptual information(such as texture)and other high-level semantic information(such as human action in video).Furthermore,the analysis of video text can be used to monitor illegal videos.The key technique for extracting video text is to find,verify,and recognize video text in various languages and fonts against complex backgrounds.In this paper,we propose a novel method that combines a corner response feature map and transferred deep convolutional neural networks for detecting and recognizing video text.First,we use a corner response feature map to detect candidate text regions with a high recall.Next,we partition the candidate text regions into candidate text lines by projection analysis using two alternative methods.We then construct classification networks transferred from VGG16,ResNet50,and InceptionV3 to eliminate false positives.Finally,we develop a novel fuzzy c-means clustering-based separation algorithm to obtain a clean text layer from complex backgrounds so that the text is correctly recognized by commercial optical character recognition software.The proposed method has good performance and robustness on video text detection and recognition,which was evaluated on three publicly available test data sets and on the high-resolution test data set we constructed.