Scene text recognition brings various new challenges occurs in recent years. Detecting and recognizing textin scenes entails some of the equivalent problems as document processing, but there are also numerousnovel problems to face for recognizing text in natural scene images. Recent research in these regions hasexposed several promise but present is motionless much effort to be entire in these regions. Most existingtechniques have focused on detecting horizontal or near-horizontal texts. In this paper, we propose a newscheme which detects texts of arbitrary directions in natural scene images. Our algorithm is equipped withtwo sets of characteristics specially designed for capturing both the natural characteristics of texts usingMSER regions using Otsu method. To better estimate our algorithm and compare it with other existingalgorithms, we are using existing MSRA Dataset, ICDAR Dataset, and our new dataset, which includesvarious texts in various real-world situations. Experiments results on these standard datasets and theproposed dataset shows that our algorithm compares positively with the modern algorithms when usinghorizontal texts and accomplishes significantly improved performance on texts of random orientations incomposite natural scenes images.
展开▼
