IP protection

1 Please provide a short description of the state-of-the-art and/or current trends in the field? How does the result fit into it?
The training of people, who are deaf or have different degree of deafness, based on some of the up to date means for video communications is a problem on which significant experience is already accumulated and investigations are carried out in many countries. One of the most suitable means is the Internet, which permits their distance learning to be done without live sign language interpretation. Most of the systems, used to transfer visual information for distant training of hearing impaired people via Internet, are based on the video compression standards Н.261 and MPEG-4. In most cases the transmitted images are video sequences, which contain the image of the sign language interpreter. The systems used for transmission of moving images of this kind in accordance with the mentioned standards, are usually based on the principle for inter-frame prediction with movements’ compensation. The general disadvantage of such systems is that in result of the compression the quality of the fast moving objects (in this particular case, the hands of the sign language interpreter) is significantly deteriorated (blurred) and this decreases the comprehensibility of the represented signs.   In order to avoid these dynamic distortions is better to compress the video sequences of moving images with intra-frame compression, in correspondence with the M-JPEG standard. In this case the quality is improved, but the compression ratio is low, which requires extremely large files to be transmitted and correspondingly – received. The distance learning tools, aimed at people who are deaf, require the development of new, more efficient algorithms for sign language interpretations coding. In the new approach, presented here, instead of the usual color video are used the interpreters’ grayscale contour images obtained from the consecutive TV frames and arranged in similar way. The contour images are compressed with lossless intra-frame compression. At the receiving side the compressed data is decoded and the video sequences – restored.  In result, the sign comprehensibility is retained; the file, obtained after the compression is relatively small and correspondingly, the time necessary to download it – shorter. This approach is based on numerous investigations, in accordance with which, for the successful understanding of the interpreted information, the use of the contour images only, is good enough. The basic requirement in this case is the extracted contours to retain the sign understandability and to be with high quality, which is of greatest importance for the face and hands of the sign language interpreter.The adaptive extraction of the objects’ contours is a problem, which could be used in various applications, such as: video clips with sign language interpretations, face recognition, etc. Significant number of algorithms have already been developed and used, most renown of which are classified as follows:

-           Methods based on image filtration with gradient and Laplacian operators;

-           Edge-fitting edge detection;

-           Morphological operators;

-           Edge tracking;

-           Hough transform, etc.

The basic disadvantage of the first group of methods is that the quality of the extracted contours is not good enough for this application. The common disadvantage of the remaining groups of methods is the relatively high computational complexity and as a sequence – the slowly processing. The volume of the visual information used for the sign language interpretations is large and this is an obstackle to embed it as additional information in the existing training courses used in distance learning.

The new approach, developed by the R&D team, solves these problems to a high degree.

2 What is the problem/need/knowledge gap that the research result is responding to?  How was it addressed before?
Most of the systems, used to transfer visual information for distance training of hearing impaired people via Internet, are based on the video compression standards Н.261 and MPEG-4. The main problem for this application is that in result of the compression the fast moving objects are blurred and the comprehensibility of the sign language interpretations is lowered. The video sign language interpretation can not be added as running additional information in the existing distance learning courses, because of their large volumes. In case that higher compression is used so that to reduce the video files, the visual quality of the interpretations is significantly deteriorated, which reduces their comprehensibility.The research result is aimed at solving these problems, together with efficient compression of the visual information. For this is developed a software tool for extracting the contour images from the sequential TV frames in the original video information. The so obtained sequence of contour images is losslessly compressed and converted into a new format. The decompression is performed by a special software player. In order to retain the comprehensibility of the represented signs, are developed a group of adaptive algorithms, aimed at preserving the main contours in the image, retaining their connectivity as much as possible. The experiments with existing video clips confirmed the expected results.
3 What is the potential for further research?
The software for preparation of learning information should automate the process as much as possible, i.e. it should be a user-friendly tool. Together with the sign language interpretations, some parts of the lessons, which contain formulas (mathematical, chemical, etc.) could be also losslessly compressed, which will solve one more problem in the training courses, when such information is presented.The method could be further developed for applications in computer games and for sending pre-prepared messages in corresponding sing language to hearing impaired people (for example – announcements for disasters: floods, tsunami, hurricanes, etc.), using the mobile communications.The comparison results of the new method with other famous methods for contours extraction is given in the attached images.                                 
4 What is the potential of the research result for synergy with other research areas either in the same or in a different discipline?
The algorithms for contours extraction could be used for processing of old handwritten documents. This will permit to extract the text and then to archive the document (after scanning and converting into electronic form) using adaptive compression: the text could be compressed losslessly, and the background – with a lossy compression. In result is obtained very efficient compression, which retains the visual quality of the documents practically unchanged.
5 What is the proposed method of IPR-protection? (patent, license, trademark etc.)
All methods for IPR protection are possible.
6 What are the steps that need to be taken in order to secure the IPR-protection? What is the cost of IPR-protection?
 The method could be protected in any possible way: patent, know-how, technical cooperation.
7 What is the expected impact of the research result? (industry, society, administration etc. and target groups of beneficiaries)
The expected impact of the research result is mainly for society – to provide tools for easier communication and access to training courses in their natural language for hearing impaired people.Significant target group is the method application for computer games which will widen the entertainment market for hearing impaired.The creation of a service, which to permit hearing impaired users to receive and understand alerting messages in their natural language, even if they are not able to read, is additional target. 
8 What is you overall assessment of the scientific maturity of the research result?
The result is scientifically mature, but it needs additional work for the preparation of user-friendly software products and for the real-time implementation.
 
KEYWORDS QUANTITATIVE ASSESSMENT (0-5).
  1 2 3 4 5
Scientific maturity       x  
Synergies       x  
State-of-the-art/innovation         x
IPR-potential         x
Bookmark the permalink. Follow any comments here with the RSS feed for this post. Post a comment or leave a trackback: Trackback URL.

Post a Comment

You must be logged in to post a comment.

Visit the other applications of the INTERVALUE Platform: R&D Repository | IP Agreements

© 2009-2010 INTERVALUE Project