Trade.Information

Product Name: Research

Product Description

Barcode Recognition CEDAR is developing algorithms to read various barcodes on mail-piece images including 1-D barcodes such as POSTNET, PLANET and UCC/EAN 128 and 2-D barcodes such as PDF-417 and Datamatrix which contain information such as ZIP code, delivery address, mail-piece ID, postage value, etc. We are also conducting theoretical studies on the machine-readability of these bar codes. The following two images are used by USPS for their IBIP program to encode various information such as destination address and postage. Datamatrix PDF-417 The next two images are used for encoding ZIP codes in different situations. Planet Postnet The last three images are used by USPS for Delivery Confirmation and other value-add services. Code 128 Code 39 Code Interleaved 2 of 5 Bioinformatics and Biomedical Information Retreival Application of machine learning and pattern recognition algorithms for gene sequences and other biomedical data. Medical Forms Analysis There is considerable potential for technology to play an increasingly greater and crucial role in medical information transfer and management, especially in the area of Emergency Medical Information Systems (EMIS). EMIS workers are often the first to interact with patients, and the first to collect vital information prior to arrival at an emergency room. By automating data collection from the PreHospital Care Reports (PCR), the information can not only be managed but disseminated more efficienly. Machine Learning Research on decision tress, artificial intelligence, neural networks, genetic algorithms. forensics research handwriting recognition address interpretation signature verification penman language models digital libraries information retrieval text processing image evaluation writer identification Handwriting Recognition Handwriting recognition concerns the conversion of the analog signal of handwriting into a digital symbolic representation. The analog signal can be in the form of either a two-dimensional scanned image of paper or a temporal one-dimensional signal captured from a device such as a tablet or PDA. The principal focus at CEDAR has been on the more-difficult two-dimensional form of handwriting. This task can also be referred to as optical handwriting recognition(OHR) analogous to optical character recogntion (OCR). The OHR task is also referred to as off-line handwriting recognition or static handwriting recogntion in contrast to the one-dimensional case which is referred to as on-line or dynamic. Research on handwriting recognition has been pursued for over two decades at CEDAR. These began as research in recognition of handwritten characters and words. Due to the complexity of the task it was found necessary to focus on applications where the linguistic constraints were strong. Research at CEDAR was initially driven by application to the postal domain. Today forensic, medical and other applications provide a focus where the use of domain knowledge and contextual information make higher levels of performance possible. Several end-to-end OHR systems are under development. They are in various stages of general usability. Some of the systems are: CEDAR-FOX, for forensic examination, CEDARABIC, for Arabic writing, CEDARATER, for assessment of school writing, etc.

Company Details

Given a valid U.S. postal address, this server attempts to rewrite the address in the proper format along with the ZIP+4 code. If it is successful, you can retrieve a Postscript or a GIF file of the address for printing, with a barcode! You can... more