QR Code因其快速解碼及高資訊容量的特性被廣泛使用在我們的生活中，但綜觀現今市面上的QR Code都以黑白兩色所組成，並不具有視覺性及美術價值，使用者只能透過手機掃碼後得知該碼所要表達的內容。本論文將就QR Code提出美化的方法，配合其RS Code編碼性質來進行圖像的嵌入，並對原先他篇論文提出的嵌入方法做改良，以及在手機上實作該碼的辨識，藉由改寫QR Code的標準解碼器來解決無法在實際環境中解碼的問題，讓我們的高度視覺性QR Code具有更高的視覺性及商業價值
首先編碼部份我們將具有大標誌性質的QR Code又稱基線QR Code做小部分的更動，接著挑選QR Code的視覺區域，經同位檢測矩陣來編出合法的RS Code，獲得可由標準解碼器解碼的基線QR Code，而由於解碼端需讀取資料長度因此會對圖像做一些處裡，最後將彩色圖像嵌入即完成高度視覺性QR Code的編碼。解碼部份則使用開源的掃碼資料庫來進行的改寫，將拍攝後圖像經解碼器的徵測及校正後，讀取資料長度來建立一個遮罩好讓我們還原校正完的QR Code至嵌入前的基線QR Code，最後將基線QR Code透過標準解碼器做解碼即可獲得其中的資訊。

QR Code is widely used in our daily life because of its fast decoding and high information capacity. The QR Code on the market today is composed of black and white modules, and it does not have visual and artistic value. Users can only know the content of the code after scanning it on the mobile phone. In this paper, we will propose a beautification method for QR Code, and embed the image through its RS Code encoding property. Also, we will improve the embedding method proposed in the previous paper, and implement the identification of the code on the mobile phone. By rewriting the standard decoder of the QR Code to solve the problem of being unable to decode in the actual environment, so that our highly visual QR Code has higher visibility and commercial value.

In the encoding part, we make a small part of the change to the QR Code with a large logo image, which can be called the baseline QR Code. Next, we select the visual area of the QR Code and compile the legal RS Code through the parity check matrix to obtain a baseline QR Code that can be decoded by a standard decoder. After that, because the decoder needs to read the data length, it will do some processing on the image. And finally, the color image is embedded to complete QR Code with high visual comprehensibility encoding.

In the decoding part, we modified the algorithm of part of the open-source bar code scanner database. After the captured image is decoded and corrected by the decoder, the data length will be read to create a mask. This mask is for us to restore the corrected QR Code to the baseline QR Code, and finally decoding the baseline QR Code is performed by a standard decoder to obtain the information.

[1]Y.-H. Lin, Y.-P. Chang, and J.-L. Wu, “Appearance-based qr code beautifier,” IEEE Trans. Multimedia, vol. 15, no. 8, pp. 2198–2207, Dec. 2013.
[2]陳亭君，“內嵌無失真標誌之高度視覺性QR碼,” 國立台灣科技大學電子工程所，2017
[3]S.-S. Lin, M.-C. Hu, C.-H. Lee, and T.-Y. Lee, “Efficient qr code beautification with high quality visual Content,” IEEE Trans. Multimedia, vol. 17, no. 9 , pp. 1515–1524, Sept. 2015.
[4]魏子傑，“高度視覺性之加密雙層QR碼,” 國立台灣科技大學電子工程所，2018
[5]Information technology — Automatic identification and data capture techniques — QR Code bar code symbology specifivation, ISO/IEC Standard 18004:2015(E), 2015
[6]Wolberg, G. (1990), “Digital Image Warping,” 1^sted, 54–56, IEEE Computer Society Press.
[7]Sean Owen, Daniel Switkin, ZXing Team, Barcode Scanner(aka ZXing or Zebra Crossing), Sep. 2018 [Online]. Available: https://github.com/zxing/zxing, Accessed on: Jan. 2020
[8]R. Cox, “Qart codes,” Apr. 2012 [Online]. Available: http://research.swtch.com/qart, Accessed on: Jan. 2020
[9]H.-K. Chu, C.-S. Chang, R.-R. Lee, and N. J. Mitra, “Halftone qr codes,” ACM Trans. Graph., vol. 32, no. 6, pp. 217:1–217:8, Nov. 2013.
[10]L. Li, J. –X. Qiu, J. -F. Lu, and C.-C. Chang, “An aesthetic qr code solution based on error correction mechanism,” ELSEVIER Journal of Systems and Software, vol. 116, pp. 85–94, June 2016.
[11]陳怡靜, “貼於平面及柱面上之帶有大標誌的QR碼之編解碼,” 國立台灣科技大學電子工程研究所, 2015。