The Recognization Of Barcode And Its Application In Product Line

With the rapid development of computer technology and the improvement of image acquisition device performance, barcode recognition based on digital image processing is now gradually popularized on. The barcode recognition technology mainly focus on the study of still barcode image or the high quality barcode image, less involved in research on recognition of the motion blurred barcode image. This thesis comes from the project of the barcode recognition on production line; try to recognize the data information of motion blurred barcode image obtained by the ordinary industrial camera based on digital image processing. This paper has finished the following contents:(1)This thesis focuses on the study of the blurred image restoration, finding the reasons that cause image blurring then building the corresponding restore model. Firstly, the direction derivation method is used to identify the direction of motion blur, then the extent of motion is identified by calculating the autocorrelation of derivation image according to the features of uniform linear motion, last the most approximate restore function is constructed.(2)Combined with general barcode image recognition processing steps, using the median filter denoises the restored barcode image. This filter can effectively erase the noise while remain the clear barcode border. Using the Otsu thresholding method achieves the barcode image binarizing, which can clearly distinct the target portion and the background portion. Using Hough transform method to detect the arbitrary angle of tilted barcode image, and then correct it. Aimed at the changing feature of black bar and white blank in the barcode area to achieve the positioning and segmentation.(3)This thesis puts forward a decimal system based on binary system for representing numbers of image. Different form traditional distance of similar line unitary method, this paper invents the unitary decimalize method for EAN-13image identification. This method can achieve the discriminating of barcode numbers more simple and efficient, and it also can be applied into the recognition of other one-dimensional barcode image.(4) In this thesis we design and build a product line barcode image capture platform, which is composed of a mechanical platform of production line and the barcode image acquisition device. In this platform different blur levels of barcode image can be obtained through controlling the speed of the product line. Using the software platform Matlab to compile the human-computer interaction interface program which can achieve the online recognition of motion blurred barcode image in a limited speed.At last, the content and the method in this thesis make a useful attempt to achieve the online barcode recognizing of product barcode image when the speed is limited. At the same time the method is also useful for the still barcode image recognizing.