The Design And Implemention Of QR Code Decoding System

The Barcode is a machine-readable representation of data which has been widelyused as an approach for automatic information acquisition. The two-dimensionalbarcode, which contains information in both two directions, is an evolving kind ofbarcode with greater capacity than general one-dimensional code. QR code is onetype of matrix two-dimensional code. Due to its ability to encode Chinese characters,It is popular in China.Barcode identification is an important procedure in barcode application. Theimage based barcode scanner is currently the mainstream of variety scanners. In thispaper, an image-input QR code identification algorithm processing is designed. Thealgorithm restores the original text information by going through two procedures–image processing procedure, and data processing procedure. The image processingprocedure includes image pre-processing, QR code detection, correction andsampling. And data processing procedure includes data partition, error correction anddata decoding.As to QR code detection algorithm,3methods are used in this paper. Firstly, thecommon used method which is based on the finder pattern detection is discussed.Then two other methods based on gradient and contour features of QR code areproposed. In the end of the section, the comparison of the3methods is done. And inthe aspect of error correction algorithm, the Reed-Solomon encoding theory isdiscussed, and the decoding method and steps are introduced and implemented.The QR code identification algorithm is then implemented on an embeddedplatform with digital signal processor ADSP-BF527as the central processing unit.The system captures the image through a gray-level digital camera, detect and decodethe QR code from the image, and send the original image and text result of QR codeto PC by Ethernet interface. In this paper, both the hardware and software areintroduced. The system frame and platform are discussed, and the camera, ethernetinterface are designed. And the software of the system is combined of severalsperated modules, including the QR identification algorithm, the camera and EthernetMAC drivers, the UDP protocol stack, and the library for general image processing.Lastly, the software in PC to receive and display the image and result isimplemented for system test. The system is robust to polluted QR code, unevenillumination, blurred image, or slightly distorted, etc. The test result shows the qualified performance of the system, and proves the algorithm to be effective.