第2期 雷芳，等：金属腐蚀区域图像增强算法研究 ·391· 分量，并对亮度分量中的腐蚀细节进行多尺度增 width preserving and low-illumination image enhance- 强，以及同态滤波来改善图像光照不均的影响， ment[J].Journal of image and graphics,2017,22(10): 从而实现对金属腐蚀区域图像的增强。本文算法 1356-1363 虽应用了小波多尺度变换与同态滤波的理论，但 [8]SINGH K,KAPOOR R.Image enhancement using expos- 不是同已有文献一样将小波变换替代同态滤波里 ure based sub image histogram equalization[J].Pattern re- 的傅里叶变换，而是先对图像进行多尺度细节增 cognition letters,2014,36:10-14. [9]汪林林，余梅，安超.模糊多尺度Retinex彩色图像增强 强重构然后再应用分块同态滤波。在不改变图像 色度和饱和度分量的基础上，本文算法相比其它 [).计算机工程与应用，2012,48(7)：1343-1346. WANG Linlin,YU Mei.AN Chao.Color image enhance- 算法具有更好的色彩保真度和细节对比度，能有 ment based on fuzzy multiple-scale Retinex[J].Computer 效的用于金属腐蚀区域图像的增强，对金属腐蚀 engineering and applications,2012,48(7):1343-1346. 产物的进一步研究分析具有重要的意义。 [10]汪荣贵，朱静，杨万挺，等.基于照度分割的局部多尺度 由于本文算法需要进行多尺度细节增强与分 Retinex算法U.电子学报，2010,38(5)：1181-1186. 块同态滤波两步处理，使得算法的时间开销比较 WANG Ronggui,ZHU Jing,YANG Wanting,et al.An 大，因此本文下一步的工作主要为提高算法的实 improved local multi-scale retinex algorithm based on il- 时性。 luminance image segmentation[J.Acta electronica sinica, 参考文献： 2010,38(5):1181-1186. [1l]汪荣贵，周良，张新龙，等.基于Retinex的PEG图像增 [1]MEDEIROS F N S,RAMALHO G L B,BENTO M P,et 强新方法).中国图象图形学报，2011,16(12少：2124 al.On the evaluation of texture and color features for 2132. nondestructive corrosion detection[J].EURASIP journal WANG Ronggui,ZHOU Liang,ZHANG Xinlong,et al. on advances in signal processing,2010,2010:817473. Novel JPEG image enhancement method based on Ret- [2]纪钢，彭丽丽，王平，基于腐蚀产物颜色分析的材料腐蚀 inex theory[J].Journal of image and graphics,2011, 程度评定方法[】.重庆理工大学学报（自然科学版）， 16(12):2124-2132 2012,26(7):69-73 [12]XIAO Limei,LI Ce,WU Zongze,et al.An enhancement JI Gang,PENG Lili,WANG Ping.Research on the meth- method for X-ray image via fuzzy noise removal and ho- od for evaluation of material corrosion degree based on momorphic filtering[J].Neurocomputing,2016,195: color analysis of corrosion products[J].Journal of 56-64. Chongqing University of Technology(natural science), [13]张亚飞，谢明鸿.基于分块DCT同态滤波的彩色图像 2012,26(7):69-73 增强算法.计算机工程与设计，2013,34(5)：1752 [3]IDRIS S A,JAFER F A.Image enhancement filter evalu- 1756. ation on corrosion visual inspection[M]//SULAIMAN H A. ZHANG Yafei,XIE Minghong.Block-DCT based homo- OTHMAN M A.OTHMAN M F I.et al.Advanced Com- morphic filtering algorithm for color image enhancement puter and Communication Engineering Technology.Cham: [J].Computer engineering and design,2013,34(5): Springer,2015:651-659 1752-1756. [4]HE Kaiming,SUN Jian,TANG Xiaoou.Guided image fil- [14]马金祥，范新南，吴志祥，等.暗通道先验的大坝水下裂 tering[J].IEEE transactions on pattern analysis and ma- 缝图像增强算法.中国图象图形学报，2016,21(12)： chine intelligence,2013,35(6):1397-1409. 1574-1584 [5]PIDAPARTI R M.HINDERLITER B.MASKEY D.Eval- MA Jinxiang,FAN Xinnan,WU Zhixiang,et al.Under- uation of corrosion growth on SS304 based on textural and water dam crack image enhancement algorithm based on color features from image analysis[J].ISRN corrosion, improved dark channel prior[J].Journal of image and 2013,2013:376823 graphics,2016,21(12:1574-1584 [6]ROBERGE P R.Corrosion inspection and monitoring[M]. [15]王建新，张有会，王志巍，等.基于H$I颜色空间的单幅 Hoboken,N.J.:John Wiley Sons,2007:1-50. 图像去雾算法[J].计算机应用，2014,34(10)：2990 [7]毛东月，谢正祥，贺向前，等.自适应双向保带宽对数变 2995. 换及低照度图像增强[J1.中国图象图形学报，2017 WANG Jianxin,ZHANG Youhui,WANG Zhiwei,et al. 22(10):1356-1363. Single image defogging algorithm based on HSI color MAO Dongyue,XIE Zhengxiang,HE Xiangqian,et al. space[].Journal of computer applications,2014,34(10): Adaptive bilateral logarithm transformation with band- 2990-2995分量，并对亮度分量中的腐蚀细节进行多尺度增 强，以及同态滤波来改善图像光照不均的影响， 从而实现对金属腐蚀区域图像的增强。本文算法 虽应用了小波多尺度变换与同态滤波的理论，但 不是同已有文献一样将小波变换替代同态滤波里 的傅里叶变换，而是先对图像进行多尺度细节增 强重构然后再应用分块同态滤波。在不改变图像 色度和饱和度分量的基础上，本文算法相比其它 算法具有更好的色彩保真度和细节对比度，能有 效的用于金属腐蚀区域图像的增强，对金属腐蚀 产物的进一步研究分析具有重要的意义。 由于本文算法需要进行多尺度细节增强与分 块同态滤波两步处理，使得算法的时间开销比较 大，因此本文下一步的工作主要为提高算法的实 时性。 参考文献： MEDEIROS F N S, RAMALHO G L B, BENTO M P, et al. On the evaluation of texture and color features for nondestructive corrosion detection[J]. EURASIP journal on advances in signal processing, 2010, 2010: 817473. [1] 纪钢, 彭丽丽, 王平. 基于腐蚀产物颜色分析的材料腐蚀 程度评定方法[J]. 重庆理工大学学报 (自然科学版), 2012, 26(7): 69–73. JI Gang, PENG Lili, WANG Ping. Research on the meth￾od for evaluation of material corrosion degree based on color analysis of corrosion products[J]. Journal of Chongqing University of Technology (natural science), 2012, 26(7): 69–73. [2] IDRIS S A, JAFER F A. Image enhancement filter evalu￾ation on corrosion visual inspection[M]//SULAIMAN H A, OTHMAN M A, OTHMAN M F I, et al. Advanced Com￾puter and Communication Engineering Technology. Cham: Springer, 2015: 651–659. [3] HE Kaiming, SUN Jian, TANG Xiaoou. Guided image fil￾tering[J]. IEEE transactions on pattern analysis and ma￾chine intelligence, 2013, 35(6): 1397–1409. [4] PIDAPARTI R M, HINDERLITER B, MASKEY D. Eval￾uation of corrosion growth on SS304 based on textural and color features from image analysis[J]. ISRN corrosion, 2013, 2013: 376823. [5] ROBERGE P R. Corrosion inspection and monitoring[M]. Hoboken, N.J.: John Wiley & Sons, 2007: 1–50. [6] 毛东月, 谢正祥, 贺向前, 等. 自适应双向保带宽对数变 换及低照度图像增强[J]. 中国图象图形学报, 2017, 22(10): 1356–1363. MAO Dongyue, XIE Zhengxiang, HE Xiangqian, et al. Adaptive bilateral logarithm transformation with band- [7] width preserving and low-illumination image enhance￾ment[J]. Journal of image and graphics, 2017, 22(10): 1356–1363. SINGH K, KAPOOR R. Image enhancement using expos￾ure based sub image histogram equalization[J]. Pattern re￾cognition letters, 2014, 36: 10–14. [8] 汪林林, 余梅, 安超. 模糊多尺度 Retinex 彩色图像增强 [J]. 计算机工程与应用, 2012, 48(7): 1343–1346. WANG Linlin, YU Mei, AN Chao. Color image enhance￾ment based on fuzzy multiple-scale Retinex[J]. Computer engineering and applications, 2012, 48(7): 1343–1346. [9] 汪荣贵, 朱静, 杨万挺, 等. 基于照度分割的局部多尺度 Retinex 算法[J]. 电子学报, 2010, 38(5): 1181–1186. WANG Ronggui, ZHU Jing, YANG Wanting, et al. An improved local multi-scale retinex algorithm based on il￾luminance image segmentation[J]. Acta electronica sinica, 2010, 38(5): 1181–1186. [10] 汪荣贵, 周良, 张新龙, 等. 基于 Retinex 的 JPEG 图像增 强新方法[J]. 中国图象图形学报, 2011, 16(12): 2124– 2132. WANG Ronggui, ZHOU Liang, ZHANG Xinlong, et al. Novel JPEG image enhancement method based on Ret￾inex theory[J]. Journal of image and graphics, 2011, 16(12): 2124–2132. [11] XIAO Limei, LI Ce, WU Zongze, et al. An enhancement method for X-ray image via fuzzy noise removal and ho￾momorphic filtering[J]. Neurocomputing, 2016, 195: 56–64. [12] 张亚飞, 谢明鸿. 基于分块 DCT 同态滤波的彩色图像 增强算法[J]. 计算机工程与设计, 2013, 34(5): 1752– 1756. ZHANG Yafei, XIE Minghong. Block-DCT based homo￾morphic filtering algorithm for color image enhancement [J]. Computer engineering and design, 2013, 34(5): 1752–1756. [13] 马金祥, 范新南, 吴志祥, 等. 暗通道先验的大坝水下裂 缝图像增强算法[J]. 中国图象图形学报, 2016, 21(12): 1574–1584. MA Jinxiang, FAN Xinnan, WU Zhixiang, et al. Under￾water dam crack image enhancement algorithm based on improved dark channel prior[J]. Journal of image and graphics, 2016, 21(12): 1574–1584. [14] 王建新, 张有会, 王志巍, 等. 基于 HSI 颜色空间的单幅 图像去雾算法[J]. 计算机应用, 2014, 34(10): 2990– 2995. WANG Jianxin, ZHANG Youhui, WANG Zhiwei, et al. Single image defogging algorithm based on HSI color space[J]. Journal of computer applications, 2014, 34(10): 2990–2995. [15] 第 2 期 雷芳，等：金属腐蚀区域图像增强算法研究 ·391·