許多自動條碼定位系統使用兩階段(two-stage)的策略，包含擷取可能條碼區域和精準定位條碼於複雜場景中。它們在擷取可能條碼區域的階段，僅透過條碼的紋理(texture)特徵，以梯度資訊預測條碼主要方位(orientation)，但容易因文字圖騰所產生過多的邊緣點而偵測錯誤。基於單純使用紋理特徵的不足，本研究加入了形狀(shape)的特徵，配合Field影像格式可消除移動糢糊和減少資料量的特點，以長短軸長度比例與統計方法尋找條碼主要方位，濾除大部分文字圖騰，以完成精準條碼定位。實驗使用EAN13碼作範例，以各種旋轉角度放置條碼，系統仍然能在Field影像格式下有效分離條碼與文字圖騰以產生條碼可能區域，並能即時偵測有無條碼與定位條碼區域。

Many automatic real-time barcode localization systems incorporate a two-stage procedure that includes possible barcode region extraction and precise barcode localization. These systems, in the stage of extracting possible barcode region, makes use of the texture features of barcode and, through gradient detection algorithm, produce gradient information that is used to determine the orientation of barcode. This method, however, tends to be erroneous whenever the source barcode is mixed with texts and graphics that contain many edges. In this research, to precisely determine the orientation of barcode and therefore achieve precise barcode localization, the shape features of barcode are taken into account alongside the texture features, and, using the length ratio of the major axis and minor axis of texts and graphics and statistics methodology, texts and graphics are effectively filtered away, thereby making precisely locating the orientation of barcode possible. To cope with the added calculation, the filed image format is used to eliminate the blurred images of moving objects and reduce the amount of data. A system was built according to the specifications described above, and sets of EAN-13 barcode, each positioned differently to ensure the validity and variety of the final result, were used for test. Using the field image format, the system effectively defined and separated from texts and graphics the possible barcode region; also, the system was able to determine the existence barcode and, if in existence, locate the barcode region in time.

[1].Kimura, N., Moriya, T., “Vision robot with rotational camera for searching ID tags,” Proceedings of SPIE - The International Society for Optical Engineering, 6813, art. no. 681302(2008).
[2].廖盛華，碩士論文，「FOUP標籤自動化檢視與辨識系統」，國立台灣科技大學，電機工程系，民國九十七年。
[3].Ando, S., Hontani, H., “Automatic visual searching and reading of barcodes in 3D scene,” Proc. IEEE Intl. Vehicle Electronics Conf., pp. 49-54(2001).
[4].Arnould, S., Awcock, G.J., Thomas, R., “Remote bar-code localisation using mathematical morphology,” Seventh International Conference on Image Processing and Its Applications, 2, pp. 642-646(1999).
[5].Chai, D., Hock, F., “Locating and decoding EAN-13 barcodes from images captured by digital cameras,” Fifth International Conference on Information, Communications and Signal Processing, art. no. 1689328, pp. 1595-1599(2005).
[6].Wachenfeld, S., Terlunen, S., Jiang, X., “Robust recognition of 1-D barcodes using camera phones,” 19th International Conference on Pattern Recognition, ICPR, art. no. 4761085(2008).
[7].Jain A.K., Chen Y., “Bar code localization using texture analysis”, Proceedings of the Second International Conference on Document Analysis and Recognition, pp. 41 – 44, Oct.(1993).
[8].Muniz R., Junco L., Otero A., “A robust software barcode reader using the Hough Transform,” Proceedings of International Conference on Information Intelligence and Systems, pp. 313 – 319, Nov. (1999).
[9].林伯諭，「以影像為基礎之即時條碼掃描系統」碩士論文，國立中正大學，嘉義，民國九十六年。
[10].蘇彥至，「條碼定位使用四向梯度運算」，碩士論文，大同大學，台北，民國九十一年。
[11].Zhang, C., Wang, J., Han, S., Yi, M., Zhang, Z., “ Automatic real-time barcode localization in complex scenes,” IEEE International Conference on Image Processing, pp. 497-500(2006).
[12].Chen, Y., Yang, Z., Bai, Z., Wu, J., “Simultaneous real-time segmentation of diversified barcode symbols in complex background,” Proceedings - The 1st International Conference on Intelligent Networks and Intelligent Systems, ICINIS, art. no. 4683280, pp. 527-530(2008).
[13].中華民國商品條碼策進會編，商品條碼應用手冊，經濟部商業司，台北，民國九十二年。
[14].杜振宇，自動辨識系統的基本知識，全華圖書，台北，民國九十六年。
[15].鍾國亮，影像處理與電腦視覺，東華書局，台北，民國九十五年。
[16].Otsu, N., “A threshold selection method from gray-level histogram,” IEEE Trans. Syst., Man, Cybern., Vol. 8, No. 1, pp. 62-66 (1978).
[17].Ammouri, S., Bilodeau, G.-A., “Face and hands detection and tracking applied to the monitoring of medication intake,” Proceedings of the 5th Canadian Conference on Computer and Robot Vision, CRV, art. no. 4562105, pp. 147-154(2008).
[18].Lumia, R., Shapiro, L., Zuniga, O., “A new connected components algorithm for virtual memory computers,” Computer Vision, Graphics and Image Processing, 22 (2), pp. 287-300(1983).
[19].Choksuriwong, A., Laurent, H., Emile, B., “A Comparative Study of Objects Invariant Descriptors,” ENSI de Bourges - Universite d'Orleans – Laboratoire Vision et Robotique - UPRES EA 2078 10 boulevard Lahitolle, 18020 Bourges Cedex, France.