A print disabled person is a person who cannot effectively read printed documents because of visual impairment. The print disability prevents a person from gaining information from printed material in the standard way and requires them to utilize alternative methods to access that information.
Schmoozer is a mobile-based audio assistance to read textual documents, which contains graphical contents such as images, tables and mathematical equations to overcome above- mentioned challenge.
The solution is capable of delivering higher accuracy to read printed material with existing technology playing a significant role in the assistive technology research area.
As illustrated by the high-level design in Fig 1.0, the system consists of four components, the Image capturing feature of printed document, the mobile application,sound output subsystem and the backend server.
The main target of this application is to read document contents for vision impaired user groups. Therefore, the application consists of autofocusing and image capturing techniques to reduce the time taken to capture the image.
The captured image is converted into a byte array before sending it to the backend server. The sending speed could be increased because it will compact the size of the file.
After uploading the captured image, graphical region identification is the foremost task of converting print disable document into digitized document. The accuracy of region identification is calculated with 1000 test images.
The accuracy of the function is in high volume. Therefore, this research outcome is able to prove that Histogram of Oriented Gradients (HOG) feature extraction with the combination of Support vector machine (SVM) algorithm is a good solution for extracting real content inside captured image.
Table 1 shows an accuracy calculation for graphical region identification algorithms which have been developed to identified graphical region. Table 2 shows the results generated by accuracy calculation for graphical image identification.Table 3 shows accuracy calculation for mathematical equation identification function. Most of the visually impaired readers are unable to read mathematical functions accurately as well as in a meaningful way.
The “Schmoozer”is designed for visually impaired community and its initial versionwas launched viathe EFC’s Disability centre for its student population usage.Further R&D of the product would continue under SLIIT assistive technology research group.
The Schmoozer is another successful yet noteworthy invention by the SLIIT initiated upon an idea brought forward by the Senior Lecturer in the Faculty of Computing, Dr. Anuradha Jayakody and his team (N.D.U. Gamage, K.W.C. Jayadewa, S.M.N.K.B. Senanayake, K.L.A.D. Udeshitha, Manique Gunaratne). Currently engaged in Assistive technology and Indoor Navigation research group attached to SLIIT.
0 comments: