Research Of QR Code Recognition Systerm Based On Image Processing

Two-dimensional bar code is a simple and cheap way for information storage, which overcomes the defects of one-dimensional bar code. The defects, belongs to one-dimensional bar code, not only definitely depend on network but also database. At the same time, it has a small information capacity as well. In contrast, Two-dimensional bar code has many benefits such as large information capacity, high reliability, good security, fast cognition ability, high rate of correction, low cost and etc. As a result,the2d bar code served as the carrier is widely used in automation information.QR code is a kind of two-dimensional barcode which is widely applied nowadays. However, the core technology is administered by foreign companies. In order to guarantee the safety of products of our country currently, the fast QR yards of effective coding and decoding, especially decoding becomes an important domestic research. Meanwhile, the traditional QR code identification system uses barcode scanners and it is more and more widely used in product logistics. Nevertheless, it couldn’t meet the remote requirements data acquisition.The significance of this thesis is to resolve the defects of the bar code scanner which can only be used for close collect. It captures long-distance bar code image though digital image acquisition devices. It uses efficient image processing methods on the acquired images and finally through the decoding algorithm to realize accurate decoding. This article is mainly as follows:Firstly, research on the detection and localization of bar code in background based on large number of documents. After that, put forward an improved barcode detection algorithm according to the sub-region and more features of the QR code.Secondly, the paper preprocesses the positioned QR code image in order to be consistent with the identification requirements of the QR code image decoding system. Summarize the foibles of QR code in the image preprocessing algorithm of predecessors’, and improve some parts of traditional QR code image preprocessing algorithm such as constrained least squares filtering, calibration graphics though hough and radon. At the same time, put forward a adaptive structuring elements used for binary morphology algorithm which based on the characteristic of QR positional detection pictureFinally, the paper studies the QR code decoding algorithm and make the key algorithm better and better, which improve the speed and accuracy of the algorithm, which includes:1) Image registration algorithm based on FFT replaces the traditional line scan algorithm.2) Improve the way of extracting coordinates of Calibration graph, study the RS, BCH error correction code in the application of QR code decoding algorithm and program it.The achievement is the decoding system development of the human-computer interaction, and development of the performance of the system after detection, which proves the effectiveness of the system.