Noise type recognition and intensity estimation based on K-nearest neighbors algorithm

doi:10.11772/j.issn.1001-9081.2019061109

Journal of Computer Applications ›› 2020, Vol. 40 ›› Issue (1): 264-270.DOI: 10.11772/j.issn.1001-9081.2019061109

• Frontier & interdisciplinary applications • Previous Articles Next Articles

Noise type recognition and intensity estimation based on K-nearest neighbors algorithm

WU Xiaoli, ZHENG Yifeng

School of Physical Science, University of Science and Technology of China, Hefei Anhui 230000, China

Received:2019-06-27 Revised:2019-08-07 Online:2019-10-08 Published:2020-01-10
Contact: 郑艺峰
Supported by:
This work is partially supported by the National Key Research and Development Program of China (2016YFA0400600, 2016YFA0400601,2016YFA0400602),the Natural Science Foundation of Anhui Province (1808085MA25).

基于K近邻算法的噪声种类识别和强度估计

吴小莉, 郑艺峰

中国科学技术大学物理学院, 合肥 230000

作者简介:吴小莉(1995-),女,江西赣州人,硕士研究生,主要研究方向:机器学习、噪声识别、信号处理;郑艺峰(1991-),男,江西鹰潭人,博士研究生,主要研究方向:等离子体数值计算、噪声研究、碰撞算法。
基金资助:
国家重点研发计划项目（2016YFA04006，2016YFA0400601，2016YFA0400602）；安徽省自然科学基金资助项目（1808085MA25）。

Abstract

Abstract: For the problem that the existing methods for noise type recognition and intensity estimation all focus on single noises, and cannot estimate the intensity of source noises in the mixed noises, a K-Nearest Neighbors (KNN) algorithm with distance threshold was proposed to recognize the single and mixed noises, and estimate the intensity of source noises in the mixed noises by combining the recognition results of mixed noises and the reconstruction of noise bases. Firstly, the data distribution in frequency domain was used as feature vector. Then the signals were identified by the noise type recognition algorithm, and the frequency domain cosine distance between reconstructed noise and real noise was adopted as the optimal evaluation criterion in the process of reconstruction of noise bases. Finally, the intensity of source noises was estimated. The experimental results on two test databases indicate that, the proposed algorithm has the average accuracy of noise type identification as high as 98.135%, and the error rate of intensity estimation of mixed noise of 20.96%. The results verify the accuracy and generalization of noise type recognition algorithm as well as the feasibility of mixed noise intensity estimation algorithm, and this method provides a new idea for the mixed noise intensity estimation. The information of mixed noise type and intensity obtained by this method contributes to the determination of denoising methods and parameters, and improves the denoising efficiency.

Key words: K-Nearest Neighbors (KNN) algorithm, distance threshold, noise base reconstruction, type recognition, intensity estimation, mixed noise

摘要： 对于目前噪声种类识别和强度估计方法都是针对单噪声，无法估计混合噪声中源噪声的强度的问题，提出了一种有距离阈值的K近邻（KNN）算法，实现对单噪声和混合噪声的种类识别，并结合混合噪声识别结果和噪声基重构估计混合噪声中源噪声的强度。首先，选用频域数据分布作为特征向量；然后，采用噪声种类识别算法进行种类识别，并且在噪声基重构过程中以重构噪声与真实噪声的频域余弦距离作为强度估计算法的最优化评价标准；最后，实现对源噪声强度的估计。在两个测试数据库上的实验结果表明，所提算法的噪声种类识别的平均精度高达98.135%，混合噪声强度估计的误差率为20.96%。实验结果验证了噪声种类识别算法的准确性和泛化性，以及混合噪声强度估计算法的可行性，并且该方法为混合噪声强度估计提供了新思路。采用该方法获取的混合噪声种类和强度信息有助于去噪方法和去噪参数的确定，进而提高去噪效率。

关键词: K近邻算法, 距离阈值, 噪声基重构, 种类识别, 强度估计, 混合噪声

CLC Number:

TP391.4

WU Xiaoli, ZHENG Yifeng. Noise type recognition and intensity estimation based on K-nearest neighbors algorithm[J]. Journal of Computer Applications, 2020, 40(1): 264-270.

吴小莉, 郑艺峰. 基于K近邻算法的噪声种类识别和强度估计[J]. 计算机应用, 2020, 40(1): 264-270.

References

[1] WEN Y, NG M K, HUANG Y. Efficient total variation minimization methods for color image restoration[J]. IEEE Transactions on Image Processing, 2008, 17(11):2081-2088.
[2] TANG Q. JIAO L. Image denoising with geometrical thresholds[J]. Electronics Letters, 2009, 45(8):405-406.
[3] PORTILLA J, STRELA V, WAINWRIGHT M J, et al. Image denoising using scale mixtures of Gaussians in the wavelet domain[J]. IEEE Transactions on Image Processing, 2003, 12(11):1338-1351.
[4] UZKENT B, BARKANA B D, YANG J. Automatic environmental noise source classification model using fuzzy logic[J]. Expert Systems with Applications, 2011, 38(7):8751-8755.
[5] AKBACAK M, HANSEN J H L. Environmental sniffing:noise knowledge estimation for robust speech systems[J]. IEEE Transactions on Audio, Speech and Language Processing, 2007, 15(2):465-477.
[6] CHOI J H, CHANG J H. On using acoustic environment classification for statistical model-based speech enhancement[J]. Speech Communication, 2012, 54(3):477-490.
[7] KARIBASAPPA K G, HIREMATH S, KARIBASAPPA K. Neural network based noise identification in digital image[J]. International Journal on Network Security, 2016, 1:28-31.
[8] KHAW H Y, SOON F C, CHUAH J H, et al. Image noise types recognition using convolutional neural network with principal components analysis[J]. IET Image Processing, 2017, 11(12):1238-1245.
[9] KUMAR S V, NAGARAJU C. FFBF:cluster-based Fuzzy Firefly Bayes Filter for noise identification and removal from grayscale images[J]. Cluster Computing, 2019, 22(S1):1289-1311.
[10] BALABANOVA I, GEORGIEV G, KOSTADINOVA S. Artificial neural network for identification of signals with superposed noises[C]//Proceedings of the 2017 International Conference on Intelligent Information Technologies for Industry, AISC 679. Cham:Springer, 2017:488-495.
[11] CAO J, CAO M, WANG J, et al. Urban noise recognition with convolutional neural network[J]. Multimedia Tools and Applications, 2019, 78(20):29021-29041.
[12] EAMDEELERD C, SONGWATANA K. Audio noise classification using bark scale features and K-NN technique[C]//Proceedings of the 2008 International Symposium on Communications and Information Technologies. Piscataway:IEEE, 2008:131-134.
[13] 袁文浩,林家骏,陈宁,等. 一种基于Bark域能量分布的噪声分类方法[J]. 华东理工大学学报(自然科学版), 2013, 39(4):472-476. (YUAN W H, LIN J J, CHEN N, et al. A noise classification method based on noise energy distribution in Bark domain[J]. Journal of East China University of Science and Technology (Natural Science Edition), 2013, 39(4):472-476.)
[14] YUAN W, XIA B. A speech enhancement approach based on noise classification[J]. Applied Acoustics, 2015, 96:11-19.
[15] 王海超,王春光,宗哲英,等. 基于噪声类型及强度估计的狭叶锦鸡儿叶切片图像盲去噪[J]. 农业工程学报, 2017, 33(10):229-238. (WANG H C, WANG C G, ZONG Z Y, et al. Blind image denoising of microscopic slices image of Caragana stenophylla Pojark based on noise type and intensity estimation[J]. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33(10):229-238.)
[16] 柳薇. SVD域的图像高斯噪声强度估计[J].中国图象图形学报, 2012, 17(8):923-933. (LIU W. Gaussian noise level estimation in SVD domain for images[J]. Journal of Image and Graphics, 2012, 17(8):923-933.)
[17] 乔双,吴晓阳,赵辰一,等. 基于PCA和BM3D的噪声估计方法及其在中子图像去噪中的应用[J]. 原子能科学技术, 2018, 52(4):729-736. (QIAO S, WU X Y, ZHAO C Y, et al. Noise level estimation method based on PCA and BM3D for neutron image denoising[J]. Atomic Energy Science and Technology, 2018, 52(4):729-736.)
[18] 吴疆,尤飞,蒋平. 基于回归分析和主成分分析的噪声方差估计方法[J]. 电子与信息学报, 2018, 40(5):1195-1201. (WU J, YOU F, JIANG P. Noise variance estimation method based on regression analysis and principal component analysis[J]. Journal of Electronics and Information Technology, 2018, 40(5):1195-1201.)
[19] 张旗,梁德群,樊鑫,等. 基于小波域的图像噪声类型识别与估计[J].红外与毫米波学报, 2004, 23(4):281-285. (ZHANG Q, LIANG D Q, FAN X, et al. Identifying of noise types and estimating of noise level for a noisy image in the wavelet domain[J]. Journal of Infrared and Millimeter Waves, 2004, 23(4):281-285.)
[20] COHEN I, BERDUGO B. Noise estimation by minima controlled recursive averaging for robust speech enhancement[J]. IEEE Signal Processing Letters, 2002, 9(1):12-15.
[21] 胡岸,高勇. 基于变窗长搜索的改进型噪声估计算法[J]. 计算机系统应用, 2018, 27(9):124-129. (HU A, GAO Y. Improved noise estimation algorithm based on searching by variable window[J]. Computer Systems and Applications, 2018, 27(9):124-129.)
[22] 马俊,姜卫平,邓连生,等. GPS坐标时间序列噪声估计及相关性分析[J]. 武汉大学学报(信息科学版), 2018, 43(10):1451-1457. (MA J, JIANG W P, DENG L S, et al. Estimation method and correlation analysis for noise in GPS coordinate time series[J]. Geomatics and Information Science of Wuhan University, 2018, 43(10):1451-1457.)
[23] 李航. 统计学习方法[M]. 北京:清华大学出版社, 2012:37. (LI H. Statistical Learning Method[M]. Beijing:Tsinghua University Press, 2012:37.)
[24] PARK H K, LUHMANN N C Jr, DONNÉ A J H, et al. Comparative study of sawtooth physics and Alfvén waves via 2-D ECE Imaging on KSTAR, DIII-D, AUG and TEXTOR[J]. Revista Chilena de Nutricion, 2012, 37(4):439-445.
[25] HAN X, LIU X, LIU Y, et al. Design and characterization of a 32-channel heterodyne radiometer for electron cyclotron emission measurements on experimental advanced superconducting tokomak[J]. Review of Scientific Instruments, 2014, 85(7):No.073506.
[26] DENG B H, BROWER D L, CIMA G, et al. Imaging of core electron temperature fluctuations of tokomak plasmas[J]. IEEE Transactions on Plasma Science, 2002, 30(1):72-73.

Noise type recognition and intensity estimation based on K-nearest neighbors algorithm

基于K近邻算法的噪声种类识别和强度估计

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 4

Recommended Articles

Metrics

[1]	LIU Bo, WANG Shengzheng, ZHAO Jiansen, LI Mingfeng. Ship tracking and recognition based on Darknet network and YOLOv3 algorithm [J]. Journal of Computer Applications, 2019, 39(6): 1663-1668.
[2]	ZHAO Jingkun, ZHOU Yingyue, LIN Maosong. Image denoising algorithm based on sparse representation and nonlocal similarity [J]. Journal of Computer Applications, 2016, 36(2): 551-555.
[3]	JIANG Lihui CHEN Hong ZHUANG Zibo XIONG Xinglong YU Lan. Wind shear recognition based on improved genetic algorithm and wavelet moment [J]. Journal of Computer Applications, 2014, 34(3): 898-901.
[4]	Hai-bo WANG Wen-hui ZHANG Hui-hua YANG Huan CHEN. Adaptive light radiation intensity estimation based on variable kernel [J]. Journal of Computer Applications, 2011, 31(08): 2240-2242.