基于區(qū)域?qū)Ρ榷仍鰪?qiáng)的二值化算法
doi: 10.11999/JEIT160197 cstr: 32379.14.JEIT160197
基金項(xiàng)目:
哈爾濱市科技創(chuàng)新人才項(xiàng)目(2014RFQXJ163)
Binarization Method Based on Local Contrast Enhancement
Funds:
The Science and Technology Innovation Talents Project of Harbin (2014RFQXJ163)
-
摘要: 降質(zhì)文檔圖像二值化問題是圖像處理領(lǐng)域的一個(gè)難點(diǎn)。該文通過分析圖像不同區(qū)域灰度對比度的差異，為降質(zhì)文檔圖像提出了新的二值化算法。首先利用四叉樹原理自適應(yīng)劃分區(qū)域，再對不同灰度對比度區(qū)域采用不同對比度增強(qiáng)法以調(diào)整局部區(qū)域內(nèi)的灰度對比度，最后根據(jù)灰度值出現(xiàn)的頻率確定局部閾值。該文測試了隨機(jī)拍攝的降質(zhì)圖像及DIBCO(Document Image Binarization COntest)圖像集中的50幅圖像。與4種經(jīng)典算法比較，所提算法處理的降質(zhì)圖像具有最高F-measure值和峰值信噪比(PSNR值)。
-
關(guān)鍵詞:
- 圖像處理 /
- 二值化 /
- 區(qū)域?qū)Ρ榷仍鰪?qiáng) /
- 局部閾值 /
- 四叉樹法
Abstract: Binarization for degraded document images is a difficult point in image processing. This paper presents a new binarization method for the degraded document images by analyzing the differences of image grayscale contrast in different areas. Firstly, theory of quadtree is used to divide areas adaptively. Secondly, various contrast enhancements are selected to adjust local grayscale contrast for different contrast areas. Lastly, the frequency of gray value is utilized to calculate threshold. The proposed algorithm is tested on random shooting degraded images and datasets of Document Image Binarization COntest (DIBCO). Compared with other four classical algorithms, the binaried images using the proposed algorithm gain the highest F-measure and PSNR (Peak Signal-to-Noise Ratio).-
Key words:
- Image processing /
- Binarization /
- Local contrast enhancement /
- Local threshold /
- Quadtree
-
OTSU N. A threshold selection method from gray level histograms[J]. IEEE Transactions on Systems, Man and Cybernetics, 1979, 9(1): 62-66. doi: 10.1109/TSMC.1979. 4310076. 申鉉京, 龍建武, 陳海鵬, 等. 三維直方圖重建和降維的Otsu閾值分割算法[J]. 電子學(xué)報(bào), 2011, 39(5): 1108-1114. SHEN Xuanjing, LONG Jianwu, CHEN Haipeng, et al. Otsu thresholding algorithm based on rebuilding and dimension reduction of the 3-dimensional histogram[J]. Acta Electronica Sinica, 2011, 39(5): 1108-1114. NIBLACK W. An Introduction to Digital Image Processing [M]. Englewood Cliffs, NJ, US, Prentice-Hall, Inc., 1986: 115-116. SAUVOLA J and PIETIKAINEN M. Adaptive document image binarization[J]. Pattern Recognit, 2000, 33(2): 225-236. doi: 10.1016/S0031-3203(99)00055-2. MA L and STAUNTON R C. A modified fuzzy C-means image segmentation algorithm for use with uneven illumination patterns[J]. Pattern Recognition, 2007, 40(11): 3005-3011. doi: 10.1016/j.patcog.2007.02.005. CHOU C H, and LIN W H, and CHANG F. A binarization method with learning-build rules for document images produced by cameras[J]. Pattern Recognition, 2010, 43(4): 1518-1530. doi: 10.1016/j.patcog.2009.10.016. 龍建武, 申鉉京, 臧慧, 等. 高斯尺度空間下估計(jì)背景的自適應(yīng)閾值分割算法[J]. 自動(dòng)化學(xué)報(bào), 2014, 40(8): 1773-1782. doi: 10.3724/SP.J.1004.2014.01773. LONG Jianwu, SHEN Xuanjing, ZANG Hui, et al. An adaptive thresholding algorithm by background estimation in Gaussian scale space[J]. Acta Automatica Sinica, 2014, 40(8): 1773-1782. doi: 10.3724/SP.J.1004.2014.01773. SINGH B M, SHARMA R, GHOSH D, et al. Adaptive binarization of severely degraded and non-uniformly illuminated documents[J]. International Journal of Document Analysis and Recognition, 2014, 17(4): 393-412. doi: 10.1007/ s10032-014-0219-6. MESQUITA R G, MELLO C A B, and ALMEIDA L H E V. A new thresholding algorithm for document images based on the perception of objects by distance[J]. Integrated Computer-Aided Engineering, 2014, 21(2): 133-146. doi: 10.3233/ICA-130453. MILYAEV S, BARINOVA O, NOVIKOVA T, et al. Fast and accurate scene text understanding with image binarization and off-the-shelf OCR[J]. International Journal of Document Analysis and Recognition, 2015, 18(2): 169-182. doi: 10.1007/ s10032-015-0240-4. ROSENFELD A and KAK A C. Digital Picture Processing [M]. 2nd ed. New York, Morgan Kaufmann: Academic Press, 1982: 92-95. -
計(jì)量
- 文章訪問數(shù): 1616
- HTML全文瀏覽量: 161
- PDF下載量: 525
- 被引次數(shù): 0
下載:
