基于火焰彩色纹理复杂度特征的转炉炼钢吹炼状态识别

doi:10.11772/j.issn.1001-9081.2015.01.0283

计算机应用 ›› 2015, Vol. 35 ›› Issue (1): 283-288.DOI: 10.11772/j.issn.1001-9081.2015.01.0283

基于火焰彩色纹理复杂度特征的转炉炼钢吹炼状态识别

李鹏举, 刘辉, 王彬, 王龙

昆明理工大学信息工程与自动化学院, 昆明650500

收稿日期:2014-07-28 修回日期:2014-09-09 出版日期:2015-01-01 发布日期:2015-01-26
通讯作者: 刘辉
作者简介:李鹏举(1990-),男,河南鹤壁人,硕士研究生,CCF会员,主要研究方向:图像处理、模式识别;刘辉(1984-),男,陕西蒲城人,讲师,博士,CCF会员,主要研究方向:实时计算机控制、图像处理、模式识别;王彬(1977-),女,黑龙江哈尔滨人,副教授,博士,CCF会员,主要研究方向:工业实时控制、模型驱动的软件设计、智能信息处理;王龙(1988-),男,黑龙江鸡西人,硕士研究生,CCF会员,主要研究方向:图像处理.
基金资助:
国家自然科学基金资助项目(61263017);云南省自然科学基金资助项目(2011FZ060, KKSY201303120).

Blowing state recognition of basic oxygen furnace based on feature of flame color texture complexity

LI Pengju, LIU Hui, WANG Bin, WANG Long

Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming Yunnan 650500, China

Received:2014-07-28 Revised:2014-09-09 Online:2015-01-01 Published:2015-01-26

摘要/Abstract

摘要：

在基于火焰图像识别的转炉吹炼状态识别过程中,针对已有方法存在火焰彩色纹理信息利用不充分和状态识别率仍需提高的问题,提出一种基于火焰彩色纹理复杂度特征的转炉吹炼状态识别方法.首先,将火焰图像转化到HSI颜色空间下并作非均匀量化;然后,计算H分量和S分量的共生矩阵从而融入火焰图像的颜色信息;其次,利用得到的颜色共生矩阵计算火焰纹理复杂度的特征描述子;最后,应用Canberra距离作为相似度度量准则对吹炼状态进行分类和识别.实验结果表明,与已有的转炉火焰灰度共生矩阵和灰度差分统计方法相比,在满足吹炼识别实时性要求的前提下,所提方法的识别率分别提高了28.33%和3.33%.

关键词: 转炉炼钢, 彩色纹理, 颜色共生矩阵, Canberra距离, 纹理识别

Abstract:

In the process of converter blowing state recognition based on flame image recognition, flame color texture information is underutilized and state recognition rate still needs to be improved in the existing methods. To deal with this problem, a new converter blowing recognition method based on feature of flame color texture complexity was proposed. Firstly, the flame image was transformed into HSI color space, and was nonuniformly quantified; secondly, the co-occurrence matrix of H component and S component was computed in order to fuse color information of flame images; thirdly, the feature descriptor of flame texture complexity was calculated using color co-occurrence matrix; finally, the Canberra distance was used as similarity criteria to classify and identify blowing state. The experimental results show that in the premise of real-time requirements, the recognition rate of the proposed method is increased by 28.33% and 3.33% respectively, compared with the methods of Gray-level co-occurrence matrix and gray differential statistics.

Key words: Basic Oxygen Furnace (BOF), color texture, color co-occurrence matrix, Canberra distance, texture recognition

中图分类号:

TP391.413

李鹏举, 刘辉, 王彬, 王龙. 基于火焰彩色纹理复杂度特征的转炉炼钢吹炼状态识别[J]. 计算机应用, 2015, 35(1): 283-288.

LI Pengju, LIU Hui, WANG Bin, WANG Long. Blowing state recognition of basic oxygen furnace based on feature of flame color texture complexity[J]. Journal of Computer Applications, 2015, 35(1): 283-288.

参考文献

[1] LIU C. Development, innovation and prospect of converter steelmaking technologies in China [J]. Special Steel Technology, 2013, 19(4):6-9.(刘超.中国转炉炼钢技术的发展、创新与展望[J].特钢技术,2013,19(4):6-9.)
[2] WANG Z, REN K, LIU Z. Sublance technology application of 120 t converter at Laiwu steelmaking plant [J]. Journal of Materials Sciences and Technology, 2009, 19(11): 30-32.(王忠刚,任科社,刘忠建.副枪技术在莱钢120 t转炉上的应用[J].中国冶金,2009,19(11):30-32.)
[3] SHI Z, WANG G, LI Y. Gas analysis in converter dynamic control technology and its application at Masteel [J]. Iron and Steel, 2007, 42(4): 24-26.(石知机,汪国才,李应江.炉气分析终点控制技术在马钢转炉的应用[J].钢铁,2007,42(4):24-26.)
[4] WEN H, ZHAO Q, CHEN Y, et al. Basic oxygen furnace endpoint forecasting model based on radiation and modified neural network [J]. Acta Optica Sinica, 2008, 28(11):2131-2135.(温宏愿,赵琦,陈延如,等.基于炉口辐射和改进神经网络的转炉终点预测模型[J].光学学报,2008,28(11):2131-2135.)
[5] LIU H, ZHANG Y, ZHANG Y, et al. State recognition of BOF based on flame image features and GRNN [J]. Computer Engineering and Applications, 2011, 47(26): 7-10.(刘辉,张云生,张印辉,等.基于火焰图像特征与GRNN的转炉吹炼状态识别[J].计算机工程与应用,2011,47(26):7-10.)
[6] XU L, LI W, ZHANG M, et al. A model of Basic Oxygen Furnace (BOF) end-point prediction based on spectrum information of the furnace flame with Support Vector Machine (SVM) [J]. Optik-International Journal for Light and Electron Optics, 2011, 122(7): 594-598.
[7] LIU H, WANG B, XIONG X. Basic oxygen furnace steelmaking end-point prediction based on computer vision and general regression neural network [J]. Optik-International Journal for Light and Electron Optics, 2014, 125(18): 5241-5248.
[8] LIU H, ZHANG Y, ZHANG Y, et al. Texture feature extraction of flame image based on gray-scale difference statistics [J]. Control Engineering of China, 2013, 20(2): 213-218.(刘辉,张云生,张印辉,等.基于灰度差分统计的火焰图像纹理特征提取[J].控制工程,2013,20(2):213-218.)
[9] GADELMAWLA E S. A vision system for surface roughness characterization using the gray level co-occurrence matrix [J]. NDT&E International, 2004, 37(7): 577-588.
[10] YOU J, WANG S, LI X, et al. Estimate blowing final point by analysing texture features of top-blowing BOF vessel mouth flame [J].Journal of University of Science and Technology Beijing, 2000, 22(6): 524-528.(尤佳,王绍纯,李希胜,等.通过转炉炉口火焰纹理分析判断氧气顶吹转炉吹炼终点[J].北京科技大学学报,2000,22(6):524-528.)
[11] MENDOZA F, VALOUS N A, ALLEN P, et al. Analysis and classification of commercial ham slice image using directional fractal dimension features [J]. Meat Science, 2009, 81(2): 313-320.
[12] PALM C. Color texture classification by integrative co-occurrence matrices [J]. Pattern Recognition, 2004, 37(5): 965-976.
[13] LIN T. A new adaptive center weighted median filter for suppressing impulsive noise in images [J]. Information Science, 2007, 177(4): 1073-1087.
[14] ZHOU L, HUANG S. Image dimensionality reduction based on HSI color model [J]. Modern Electronics Technique, 2013, 36(14): 79-81.(周刘兵,黄硕.基于HSI彩色模型的图像降维技术[J].现代电子技术,2013,36(14):79-81.)
[15] DING S, CHANG X, WU Q, et al. Comparative study of two dimensional vectors pattern classification based on GRNN and BPNN [J]. Foreign Electronic Measurement Technology, 2014, 33(5): 56-58.(丁硕,常晓恒,巫庆辉,等.基于GRNN与BPNN的二维向量模式分类对比研究[J].国外电子测量技术,2014,33(5):56-58.)
[16] WANG J, GE Y. Texture feature recognition based on Contourlet transform and support vector machine [J]. Journal of Computer Applications, 2013, 33(3): 677-679.(王佳奕,葛玉荣.基于Contourlet变换和支持向量机的纹理识别方法[J].计算机应用,2013,33(3):677-679.)
[17] YAN X. Weighted KNN classification algorithm based on mean distance of category [J]. Computer Systems and Applications, 2014, 23(2): 128-132.(严晓明.基于类别平均距离的加权KNN分类算法[J].计算机系统应用,2014,23(2):128-132.)
[18] HUANG Y, WANG Y. Decision tree classification based on naive Bayesian and ID3 algorithm [J]. Computer Engineering, 2012, 38(14): 41-43.(黄宇达,王迤冉.基于朴素贝叶斯与ID3算法的决策树分类[J].计算机工程,2012,38(14):41-43.)
[19] FAKHERI M, SEDGHI T, SHAYESTEH M G, et al. Framework for image retrieval using machine learning and statistical similarity matching techniques [J]. IET Image Processing, 2013, 7(1): 1-11.
[20] LIU G, YANG J. Content-based image retrieval using color difference histogram [J]. Pattern Recognition, 2013, 46(1): 188-198.

基于火焰彩色纹理复杂度特征的转炉炼钢吹炼状态识别

Blowing state recognition of basic oxygen furnace based on feature of flame color texture complexity

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

[1]	蒋考林, 白玮, 张磊, 陈军, 潘志松, 郭世泽. 基于多通道图像深度学习的恶意代码检测[J]. 计算机应用, 2021, 41(4): 1142-1147.
[2]	王佳奕葛玉荣. 基于Contourlet变换和支持向量机的纹理识别方法[J]. 计算机应用, 2013, 33(03): 677-679.