An Accessible QR Code Scanning Software Based On Image Recognition For Mobile Device

With the popularity of smart devices and mobile Internet, two-dimensional code technology, especially the QR code, because of its unique advantages of fast response speed, large information capacity, etc, has been widely used. People can obtain valid information so easily by scanning the two-dimensional codes using scanning softwares. According to the CDPF in 2010, the number of visually impaired people is about 12.63 million, and is increasing year by year, there is no doubt that the visually impaired people has been a group that mobile Internet can’t ignore. Visually impaired people due to a lack of visual ability, can only rely more on hearing or touch to use mobile devices. However, through the investigation on the market for mobile phone QR code scanning softwares, I have found that almost no one makes some optimization for the visually impaired people, these softwares is inefficient and not so friendly to scan two-dimensional codes for those visually impaired group. Therefore, the development of a barrier-free dimensional code scanning software for the visually impaired people is of great significance. In this context, our laboratory cooperates with China Braille Press and hopes to develop a two-dimensional code scanning software accessible to the visually impaired people.According to the characteristics of the visually impaired people, this paper proposes a remind algorithm based on the location of QR code local orientation locator. The algorithm extracts the possible QR code from the image by identifying the compositional ratio of pixels of QR code unique position locator. In order to apply the localization algorithm to smart mobile devices and improve the scanning efficiency, the paper adopted a multi-level graphics filter algorithm, and different voice tips can be given in real time according to the results of detection. At the same time the subject project improves the code binary arithmetic process on the basis of the two-dimensional code scanning third-party libraries Zxing from Google open source code based on IOS system, using local dynamic threshold selection method instead of the original global threshold selection method, thus makes it possible that the code has a higher processing efficiency on uneven illumination image. Secondly, the paper designs and implements the accessibility QR code scanning software on IOS system platform by Xcode and Objective-C language, and adds human intelligent voice reminder for the software, when the scan is successful, results of the scan will be read, and You can quickly browse and share scan information to help visually impaired people to complete the scan operation faster and more accurately. Finally, the paper tests the performance of the software and other mainstream softwares on the real machine equipment and compares two different performance results which demonstrates the feasibility of our software in improving the scanning efficiency and experience for the visually impaired people.