摘要
金相图像是金属材料微观形貌的体现,对金相图像进行分析是研究金属材料的重要手段。针对金相图像噪声强、晶界弱、光照不均匀等特点,结合数字图像处理方法,提出了一种金相图像二值降噪方法。首先,采用一种预处理综合方案对图像进行均衡处理。然后采取自适应阈值的方法得到金相二值图像,但仍存在较多噪声。对此,提出一种基于多尺度模板的降噪方法,该方法能有效去除大部分噪声。结果表明,该方法能够得到较好的图像预处理结果和清晰且较为准确的二值化结果,这将有利于后续金相定量分析工作的开展。
The metallographic image was the embodiment of the metal surface morphology.The analysis of the metallographic image was an important means to study the metal material.Aiming at the characteristics of metallographic image noise,weak grain boundary and uneven illumination,combined with digital image processing method,a binary image denoising method for metallographic image was proposed.First of all,a pre-processing integrated scheme to equalize the image was carried out.The adaptive threshold method was then used to obtain the metallographic binary image,but there was still more noise.In this regard,a multi-scale template based noise reduction method was proposed,which can effectively remove most of the noise.The results show that the proposed method can obtain better image preprocessing results and clear and accurate binarization results,which will be beneficial to the subsequent metallographic quantitative analysis.
引文
[1]王大海.基于图像处理技术的金相分析研究[J].铸造技术,201435(7):1619-1622.
[2]Lei Li,Xiaogang Li,Chaofang Dong,Yizhong Huang.Computational simulation of metastable pitting of stainless steel[J].Electrochimica Acta,2009,54(26):6389-6395.
[3]靳伍银,赵霞霞,祁晓玲,等.基于数字图像处理的GCr15轴承钢金相组织定量分析[J].兰州理工大学学报,2013,39(1):6-9.
[4]朱钱兵,张覃轶.金属材料晶粒度数字评定方法及应用[J].金属热处理,2014,39(9):132-136.
[5]Wang H,Zhang Y.A Detail Enhancement Algorithm of Infrared Images Based on Second-genera tion Wavelet Cumulative Probability[C].Asian Pacific Conference on Energy,Environment and Sustainable Development.2017:199-202.
[6]K.Kokufuta,T.Maruyanm.Real-time processing of local contrast enhancement on FPGA[C].International conference on field programmable logic and applications.2009:288-293.
[7]徐黎明,吕继东.基于同态滤波和K均值聚类算法的杨梅图像分割[J].农业工程学报,2015,31(14):202-208.
[8]He K,Sun J,Tang X.Guided image filtering[J].IEEE transactions on pattern analysis and machine intelligence,2013,35(6):1397-1409.
[9]温锐,陈宏文,张雷,等.基于引导滤波的多图谱医学图像分割[J].南方医科大学学报,2015,35(9):1263-1267.
[10]陈勇,帅锋,樊强.基于自然统计特征分布的无参考图像质量评价[J].电子与信息学报,2016,38(7):1645-1653.
[11]周莉莉,姜枫.图像分割方法综述研究[J].计算机应用研究,2017,34(7):1921-1928.