万英文字符,标志检测识别.docVIP

Automatic Document Logo Detection Guangyu Zhu and David Doermann Institute for Advanced Computer Studies University of Maryland, College Park, MD 20742, USA { zhugy, doermann} @ Abstract Automatic logo detection and recognition continues to be of great interest to the document retrieval community as it enables effective identification of the source of a document. In this paper, we propose a new approach to logo detection and extraction in document images that robustly classifies and precisely localizes logos using a boosting strategy across multiple image scales. At a coarse scale, a trained Fisher classifier performs initial classification using features from document context and connected components. Each logo candidate region is further classified at successively finer scales by a cascade of simple classifiers, which allows false alarms to be discarded and the detected region to be refined. Our approach is segmentation free and layout independent. We define a meaningful evaluation metric to measure the quality of logo detection using labeled ground truth. We demonstrate the effectiveness of our approach using a large collection of real-world documents. 1. Introduction a declaration of document source and ownership. The problem of logo detection and recognition is of great interest in the document domain as it enables us to identify the source of documents based on the organization where a document originates. Facing continually increasing volumes of documents, logo recognition has evolved as a practical and reliable supplement to the recognition of printed text using OCR and analysis of the textual content through natural language processing. In the context of document image retrieval, logos provide an important form of indexing that enables effective exploration of data. Given a large collection of documents, searching for a specific logo is a highly effective way of retrieving documents from the associated organization. The ability to robustly d