改進式高容量彩色二維條碼之研究

Abstract

早期二維條碼始於工業生產的零件追蹤，本身是為了有較大資訊承載量而設計，僅需可快速生產並單純讀取，因此一般可見多為黑白。伴隨科技日新月異及電腦網路蓬勃發展，各類資訊漸漸改由數位方式傳播，相較於其他二維條碼，Quick Response Code具有多樣的容錯編碼機制，又因其可快速讀取、承載較大資訊、成本低廉等特性，使得QR碼目前正大量應用於日常生活中；隨後為了更美觀及提高讀碼意願，QR碼由黑白條碼進階為圖像化二維條碼；而當需要更高容量數據時，利用色彩是有效的方法。本研究採取傳統QR碼之結構，使用色彩空間(R、G、B)分層方式增加數據容量，接著再以資訊隱藏方式在表層植入資訊點改進彩色QR碼，達到可讀取以及更多層之高容量彩色QR碼，突破以往不能以手機APP直接讀取的現況。為進一步測試解碼效果，續將改進後之高容量彩色QR碼列印輸出成五個不同大小尺寸，分別掃描並進行資料判讀及辨識錯誤分析，當條碼輸出之尺寸愈小，分層判讀之錯誤率會愈高。另採用色彩校正機制，可大幅降低錯誤率。Two-dimensional barcodes started with the inventory tracking during the production process in the manufacturing. They were designed to carry a large amount of data and simply composed of by a variety of black and white squares, which could be rapidly created and easily readable by devices. With the booming of the Internet and technology, all kinds of information are disseminated in the digital way. That is, the usage of two-dimensional barcodes becomes common. However, compared with two-dimensional barcodes, Quick Response Code has gained great popularity over the former one because of the following advantages including fast response, high accuracy, large data capacity, and reasonable cost. Thus, the usage of QR code has increased and widely applied in daily life. Subsequently, in order to improve its exterior design as well as enhance the willingness of public to use the code, the QR code has advanced from the patterns with single color to an image-based matrix, which is also known as graphic QR code. When a product contains lots of information that requires more than one code to hold, using color is the most effective method. It can hide more mechanism and bear higher data capacity. This research adopts the structure of the traditional QR code, but uses RGB color space to hide the relevant information and increase the data capacity. Meanwhile, we advance the original QR code by setting one explicit code and three internal codes in it by using the hiding mechanism. The four layers of codes will greatly strengthen the effect of anti-counterfeiting. Furthermore, in order to learn the best decoding condition, we have 5 kinds of print-and-scan size for analyzing. The smaller the size, the higher the error rate of each layer. However, with the improvement of color correction, it can increase decoding rate significantly. The outcome not only can be adopted in various application but also expand the added value of products.