《计算机应用》唯一官方网站 ›› 2024, Vol. 44 ›› Issue (11): 3565-3573.DOI: 10.11772/j.issn.1001-9081.2023111639
收稿日期:
2023-12-01
修回日期:
2024-04-16
接受日期:
2024-04-18
发布日期:
2024-05-06
出版日期:
2024-11-10
通讯作者:
邹耀斌
作者简介:
张彬(1997—),男,湖北孝感人,硕士研究生,主要研究方向:数字图像处理。
基金资助:
Yaobin ZOU1(), Bin ZHANG1,2
Received:
2023-12-01
Revised:
2024-04-16
Accepted:
2024-04-18
Online:
2024-05-06
Published:
2024-11-10
Contact:
Yaobin ZOU
About author:
ZHANG Bin, born in 1997, M. S. candidate. His research interests include digital image processing.
Supported by:
摘要:
灰度图像的灰度直方图可以呈现出无峰、单峰、双峰或多峰的形态特征,但传统熵阈值分割方法大多仅适合处理具有单峰或双峰形态特征的灰度图像。为了提高熵阈值分割方法的分割精度和分割适应性,提出一种四向加权香农熵最大化导向的自动阈值分割方法FWSE(Four-directional Weighted Shannon Entropy)。首先用新设计的方向性Prewitt卷积核在4个方向进行多尺度乘积变换(MPT),以获得一系列方向性MPT图像;再基于三次样条插值函数和曲率最大化准则自动计算出每个方向的最优MPT图像;其次在每个方向上通过内外轮廓图像对最优MPT图像的像素进行重新取样,以获取重构的灰度直方图,并在此基础上计算相应的香农熵;最后以4个方向的加权香农熵最大化为准则选取最佳分割阈值。与新近的3种阈值分割方法以及2种非阈值分割方法在4幅合成图像和100幅真实世界图像上进行实验,结果显示:在合成图像上,FWSE方法的平均马修斯相关系数(MCC)达到了0.999;在真实世界图像上,FWSE方法与其他5个分割方法的平均MCC分别是0.974、0.927、0.668、0.595、0.550和0.525。这表明FWSE方法具有更高的分割精度和更灵活的分割适应性。
中图分类号:
邹耀斌, 张彬. 四向加权香农熵最大化导向的自动阈值分割方法[J]. 计算机应用, 2024, 44(11): 3565-3573.
Yaobin ZOU, Bin ZHANG. Automatic thresholding method guided by maximizing four-directional weighted Shannon entropy[J]. Journal of Computer Applications, 2024, 44(11): 3565-3573.
分割方法 | 图5(a) | 图5(b) | 图5(c) | 图5(d) | RMCC均值 | ||||
---|---|---|---|---|---|---|---|---|---|
RMCC | 阈值 | RMCC | 阈值 | RMCC | 阈值 | RMCC | 阈值 | ||
FWSE | 1.000 | 254 | 1.000 | 169 | 0.999 | 136 | 1.000 | 154 | 0.999 |
MVET | 0.028 | 85 | 0.028 | 126 | 0.643 | 78 | 0.756 | 120 | 0.364 |
RRET | 1.000 | 254 | 0.013 | 110 | 0.892 | 95 | 0.930 | 131 | 0.709 |
EREM | 0.176 | 244 | 0.001 | 58 | 0.999 | 136 | 1.000 | 154 | 0.544 |
KLFCM | 0.039 | — | 0.084 | — | 0.905 | — | 0.839 | — | 0.467 |
GLMF | 0.021 | — | 0.012 | — | 0.912 | — | 0.900 | — | 0.461 |
表1 不同方法在合成图像上的RMCC值和分割阈值
Tab. 1 RMCC values and segmentation thresholds of different methods on synthetic images
分割方法 | 图5(a) | 图5(b) | 图5(c) | 图5(d) | RMCC均值 | ||||
---|---|---|---|---|---|---|---|---|---|
RMCC | 阈值 | RMCC | 阈值 | RMCC | 阈值 | RMCC | 阈值 | ||
FWSE | 1.000 | 254 | 1.000 | 169 | 0.999 | 136 | 1.000 | 154 | 0.999 |
MVET | 0.028 | 85 | 0.028 | 126 | 0.643 | 78 | 0.756 | 120 | 0.364 |
RRET | 1.000 | 254 | 0.013 | 110 | 0.892 | 95 | 0.930 | 131 | 0.709 |
EREM | 0.176 | 244 | 0.001 | 58 | 0.999 | 136 | 1.000 | 154 | 0.544 |
KLFCM | 0.039 | — | 0.084 | — | 0.905 | — | 0.839 | — | 0.467 |
GLMF | 0.021 | — | 0.012 | — | 0.912 | — | 0.900 | — | 0.461 |
分割 方法 | 图6(a) | 图6(b) | 图6(c) | 图6(d) | ||||
---|---|---|---|---|---|---|---|---|
RMCC | 阈值 | RMCC | 阈值 | RMCC | 阈值 | RMCC | 阈值 | |
FWSE | 0.971 | 27 | 0.963 | 167 | 0.964 | 221 | 0.948 | 135 |
MVET | 0.177 | 128 | 0.042 | 94 | 0.422 | 169 | 0.150 | 53 |
RRET | 0.210 | 112 | 0.275 | 112 | 0.148 | 144 | 0.114 | 34 |
EREM | 0.228 | 103 | 0.735 | 141 | 0.971 | 220 | 0.963 | 125 |
KLFCM | 0.207 | — | 0.038 | — | 0.780 | — | 0.148 | — |
GLMF | 0.249 | — | 0.032 | — | 0.173 | — | 0.123 | — |
表2 不同方法在4幅代表性真实世界图像上的RMCC值和分割阈值
Tab. 2 RMCC values and segmentation thresholds of different methods on four representative real-world images
分割 方法 | 图6(a) | 图6(b) | 图6(c) | 图6(d) | ||||
---|---|---|---|---|---|---|---|---|
RMCC | 阈值 | RMCC | 阈值 | RMCC | 阈值 | RMCC | 阈值 | |
FWSE | 0.971 | 27 | 0.963 | 167 | 0.964 | 221 | 0.948 | 135 |
MVET | 0.177 | 128 | 0.042 | 94 | 0.422 | 169 | 0.150 | 53 |
RRET | 0.210 | 112 | 0.275 | 112 | 0.148 | 144 | 0.114 | 34 |
EREM | 0.228 | 103 | 0.735 | 141 | 0.971 | 220 | 0.963 | 125 |
KLFCM | 0.207 | — | 0.038 | — | 0.780 | — | 0.148 | — |
GLMF | 0.249 | — | 0.032 | — | 0.173 | — | 0.123 | — |
分割方法 | 合成图像 | 真实世界图像 | ||
---|---|---|---|---|
均值 | 标准偏差 | 均值 | 标准偏差 | |
MVET | 0.000 9 | 0.000 2 | 0.001 2 | 0.000 8 |
GLMF | 0.034 9 | 0.029 7 | 0.180 0 | 0.100 6 |
EREM | 0.077 4 | 0.003 6 | 0.113 6 | 0.035 0 |
FWSE | 0.153 8 | 0.019 0 | 0.214 0 | 0.070 2 |
KLFCM | 0.125 2 | 0.057 5 | 1.180 4 | 7.038 1 |
RRET | 0.824 8 | 0.246 8 | 1.610 3 | 1.671 5 |
表3 不同方法的CPU耗时的均值和标准偏差 ( s)
Tab. 3 Mean and standard deviation of CPU running time for different methods
分割方法 | 合成图像 | 真实世界图像 | ||
---|---|---|---|---|
均值 | 标准偏差 | 均值 | 标准偏差 | |
MVET | 0.000 9 | 0.000 2 | 0.001 2 | 0.000 8 |
GLMF | 0.034 9 | 0.029 7 | 0.180 0 | 0.100 6 |
EREM | 0.077 4 | 0.003 6 | 0.113 6 | 0.035 0 |
FWSE | 0.153 8 | 0.019 0 | 0.214 0 | 0.070 2 |
KLFCM | 0.125 2 | 0.057 5 | 1.180 4 | 7.038 1 |
RRET | 0.824 8 | 0.246 8 | 1.610 3 | 1.671 5 |
1 | 曹建农.基于直方图重构的极大交叉熵图像分割方法[J]. 计算机应用,2011,31(12):3373-3377. |
CAO J N. Image segmentation of maximum cross entropy based on histogram reconstruction[J]. Journal of Computer Applications, 2011, 31(12): 3373-3377. | |
2 | YAZID H, BASAH S N, RAHIM S A, et al. Performance analysis of entropy thresholding for successful image segmentation[J]. Multimedia Tools and Applications, 2022, 81(5): 6433-6450. |
3 | 邹耀斌,齐慧康,孙水发.皮尔逊相关性最大化导向的自动阈值分割方法[J].电子测量技术,2023,46(17):109-117. |
ZOU Y B, QI H K, SUN S F. Automatic thresholding segmentation guided by maximizing Pearson correlation[J]. Electronic Measurement Technology, 2023, 46(17): 109-117. | |
4 | ASHIBA H I, ASHIBA M I. Novel proposed technique for automatic fabric defect detection[J]. Multimedia Tools and Applications, 2023, 82(20): 30783-30806. |
5 | TRUONG M T N, KIM S. Automatic image thresholding using OTSU's method and entropy weighting scheme for surface defect detection[J]. Soft Computing, 2018, 22(13): 4197-4203. |
6 | LIU C, XIE F, DONG X, et al. Small target detection from infrared remote sensing images using local adaptive thresholding[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2022, 15: 1941-1952. |
7 | KHORRAM B, YAZDI M. A new optimized thresholding method using ant colony algorithm for MR brain image segmentation[J]. Journal of Digital Imaging, 2019, 32(1): 162-174. |
8 | JIA F, SHI C, HE K, et al. Degraded document image binarization using structural symmetry of strokes[J]. Pattern Recognition, 2018, 74: 225-240. |
9 | JOSHI M, MAZUMDAR B, DEY S. Investigating the impact of thresholding and thinning methods on the performance of partial fingerprint identification systems: a review[J]. Journal of Electronic Imaging, 2023, 32(1): No.010901. |
10 | KAPUR J N, SAHOO P K, WONG A K C. A new method for gray-level picture thresholding using the entropy of the histogram[J]. Computer Vision, Graphics, and Image Processing, 1985, 29(3): 273-285. |
11 | 易三莉,张桂芳,贺建峰,等.基于最大类间方差的最大熵图像分割[J].计算机工程与科学,2018,40(10):874-1881. |
YI S L, ZHANG G F, HE J F, et al. Maximum entropy image segmentation based on maximum interclass variance[J]. Computer Engineering and Science, 2018, 40(10): 1874-1881. | |
12 | DE ALBUQUERQUE M P, ESQUEF I A, GESUALDI MELLO A R. Image thresholding using Tsallis entropy[J]. Pattern Recognition Letters, 2004, 25(9): 1059-1065. |
13 | RODRIGUES P S, GIRALDI G A. Improving the non-extensive medical image segmentation based on Tsallis entropy[J]. Pattern Analysis and Applications, 2011, 14: 369-379. |
14 | 黄聪,邹耀斌.快速二维累积剩余Tsallis熵阈值分割方法[J]. 液晶与显示,2023,38(11):1600-1614. |
HUANG C, ZOU Y B. Fast 2D cumulative residual Tsallis entropy threshold segmentation method[J]. Chinese Journal of Liquid Crystals and Displays, 2023, 38(11): 1600-1614. | |
15 | MANOHAR V, LAXMINARAYANA G, SATYA SAVITHRI T. Image compression using explored bat algorithm by Rényi 2-D histogram based on multilevel thresholding[J]. Evolutionary Intelligence, 2021, 14(1): 75-85. |
16 | SAHOO P, WILKINS C, YEAGER J. Threshold selection using Rényi's entropy[J]. Pattern Recognition, 1997, 30(1): 71-84. |
17 | 聂方彦,李建奇,张平凤,等.复杂图像的Kaniadakis熵阈值分割方法[J].激光与红外,2017,47(8):1040-1045. |
NIE F Y, LI J Q, ZHANG P F, et al. Threshold segmentation method of complex image based on Kaniadakis entropy[J]. Laser and Infrared, 2017, 47(8): 1040-1045. | |
18 | LEI B, FAN J. Infrared pedestrian segmentation algorithm based on the two-dimensional Kaniadakis entropy thresholding[J]. Knowledge-Based Systems, 2021, 225: No.107089. |
19 | NIE F, ZHANG P, LI J, et al. A novel generalized entropy and its application in image thresholding[J]. Signal Processing, 2017, 134: 23-34. |
20 | PAL N R. On minimum cross-entropy thresholding[J]. Pattern Recognition, 1996, 29(4): 575-580. |
21 | NIE F. Tsallis cross-entropy based framework for image segmentation with histogram thresholding[J]. Journal of Electronic Imaging, 2015, 24(1): No.013002. |
22 | LIN Q Q, ZHANG L, WU T L, et al. Application of Tsallis cross-entropy in image thresholding segmentation[J]. Sensors and Materials, 2020, 32(8): 2687-2696. |
23 | 范九伦,雷博.倒数粗糙熵图像阈值化分割算法[J].电子与信息学报,2020,42(1):214-221. |
FAN J L, LEI B. Image thresholding segmentation method based on reciprocal rough entropy[J]. Journal of Electronics and Information Technology, 2020, 42(1): 214-221. | |
24 | MOUSAVIRAD S J, OLIVA D, CHAKRABORTTY R K, et al. Population-based self-adaptive generalised Masi entropy for image segmentation: a novel representation[J]. Knowledge-Based Systems, 2022, 245: No.108610. |
25 | LEI B, FAN J L. Adaptive Kaniadakis entropy thresholding segmentation algorithm based on particle swarm optimization[J]. Soft Computing, 2020, 24(10): 7305-7318. |
26 | SAHOO S, SAHOO S S, KUMAR S, et al. Optimized entropy based image segmentation[C]// Proceedings of the 2020 International Conference on Computer Science, Engineering and Applications. Piscataway: IEEE, 2020: 1-6. |
27 | KANG M S, KIM K T. Automatic SAR image registration via Tsallis entropy and iterative search process[J]. IEEE Sensors Journal, 2020, 20(14): 7711-7720. |
28 | 邹耀斌,孟祥丹,孙水发,等.平稳小波域多尺度乘积下的指数Rényi熵自动阈值选择方法[J].光学精密工程,2023,31(12):1841-1858. |
ZOU Y B, MENG X D, SUN S F, et al. Automatic threshold selection method using exponential Rényi entropy under multi-scale product in stationary wavelet domain[J]. Optics and Precision Engineering, 2023, 31(12): 1841-1858. | |
29 | 邹耀斌,张进玉,臧兆祥,等.概率分布双向稀疏化下单一Tsallis熵阈值选取方法[J].软件学报,2022,33(5):1922-1946. |
ZOU Y B, ZAHNG J Y, ZANG Z X, et al. Automatic threshold selection method using single Tsallis entropy under bidirectional sparse probability distribution[J]. Journal of Software, 2022, 33(5): 1922-1946. | |
30 | 邹耀斌,邓世成,孟祥丹,等.多向加权Tsallis熵最大化导向的自动阈值分割方法[J].电子学报,2024,52(1):129-143. |
ZOU Y B, DENG S C, MENG X D, et al. Automatic thresholding segmentation method guided by maximizing multi-directional weighted Tsallis entropy[J]. Acta Electronica Sinica, 2024, 52(1): 129-143. | |
31 | 陈学松.数值分析[M].北京:机械工业出版社,2023:34-38. |
CHEN X S. Numerical analysis[M]. Beijing: China Machine Press, 2023: 34-38. | |
32 | XING J, YANG P, QINGGE L. Automatic thresholding using a modified valley emphasis[J]. IET Image Processing, 2020, 14(3): 536-544. |
33 | WANG C, PEDRYCZ W, LI Z, et al. Kullbac-Leibler divergence-based fuzzy C-means clustering incorporating morphological reconstruction and wavelet frames for image segmentation[J]. IEEE Transactions on Cybernetics, 2022, 52(8): 7612-7623. |
34 | WAN M, HUANG Q, XU Y, et al. Global and local multi-feature fusion-based active contour model for infrared image segmentation[J]. Displays, 2023, 78: No.102452. |
35 | CHICCO D, JURMAN G. The Matthews Correlation Coefficient (MCC) should replace the ROC AUC as the standard metric for assessing binary classification[J]. BioData Mining, 2023, 16: No.4. |
36 | CHICCO D, JURMAN G. The advantages of the Matthews Correlation Coefficient (MCC) over F1 score and accuracy in binary classification evaluation[J]. BMC Genomics, 2020, 21: No.6. |
37 | BENEŠ M, ZITOVÁ B. Performance evaluation of image segmentation algorithms on microscopic image data[J]. Journal of Microscopy, 2015, 257(1): 65-85. |
[1] | 贾鹤鸣, 李瑶, 姜子超, 孙康健. 基于改进共生生物搜索算法的林火图像多阈值分割[J]. 计算机应用, 2021, 41(5): 1465-1470. |
[2] | 卓禹心, 韩素雅, 张榆锋, 李支尧, 董毅峰. 基于超声谐波包络Nakagami参数图像的微波消融区域自动分割方法[J]. 计算机应用, 2021, 41(10): 3089-3096. |
[3] | 曹义亲, 刘龙标. 基于缺陷比例限制的背景差分钢轨表面缺陷检测方法[J]. 计算机应用, 2020, 40(10): 3066-3074. |
[4] | 孟令江, 王挺, 姚辰. 野外环境下图像中坑区域的提取方法[J]. 计算机应用, 2016, 36(4): 1132-1136. |
[5] | 付辉, 吴斌, 韩东轩, 黄阳强. 基于大气物理模型的快速视觉优化去雾算法[J]. 计算机应用, 2015, 35(11): 3316-3320. |
[6] | 孟若琳 邢前国. 基于可见光的海上船舶油井平台遥感检测[J]. 计算机应用, 2013, 33(03): 708-711. |
[7] | 张新明 李振云 郑颖. 融合Fisher准则和势函数的多阈值图像分割[J]. 计算机应用, 2012, 32(10): 2843-2847. |
[8] | 肖劲飞 王晓宇 陈斌 孙晓刚 刘兵. 基于K-L变换和模糊集理论的彩色字符图像分割[J]. 计算机应用, 2010, 30(9): 2464-2466. |
[9] | 郭斯羽 鲍美华 翟文娟 唐求. 内皮细胞图像自动分割与荧光强度测量方法[J]. 计算机应用, 2010, 30(12): 3274-3277. |
[10] | 杨涛 管一弘. 基于模糊Gibbs随机场聚类二维直方图的核磁共振图像分割[J]. 计算机应用, 2010, 30(10): 2797-2801. |
[11] | 张伟 隋青美. 基于小波变异粒子群和模糊熵的图像分割[J]. 计算机应用, 2010, 30(1): 54-57. |
[12] | 孙艳忠 柴毅 尹宏鹏. 基于边缘检测与改进最大类间方差法的火箭序列图像分割算法[J]. 计算机应用, 2009, 29(11): 3027-3029. |
[13] | 赵大兴 王璜 朱锦雷 颜云辉. 高精度自适应阈值分割算子的设计[J]. 计算机应用, 2008, 28(7): 1742-1743. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||