没有合适的资源?快使用搜索试试~ 我知道了~
资源推荐
资源详情
资源评论
致谢
“读书不觉已春深,一寸光阴一寸金”, 不经觉间,已轻轻放下撰写论文的
笔。揉揉略显疲倦的眼,却只是呆呆的望向窗外,心中浅浅的喜,淡淡的伤,只
觉“重回首,去年时,揽尽风雨苦亦甜”!
首先我将向我的导师周长春教授表达最真挚的谢意和崇高的敬意!“传道、
授业、解惑,师道芳香千古”,师从三载,恩师严谨的治学态度、渊博的知识、
高贵的敬业精神时时刻刻激励着我不断向前进步。恩师不仅在学术上给予我很大
的指导,亦十分关心我的生活,给予我最大的关怀!深恩永志在我心,值此机会,
再次向周长春教授表达我诚挚的谢意!
一条幽径曲折迂回、满地荆棘,走向那心旷神怡的天水一色,在潮起潮落中,
怎能忘却师兄们对我的帮助与支持!有些已分开,却总能相约“花间一壶酒”,
把酒言欢,再续情谊!感谢刘晓凯硕士、王松涛硕士、丛龙斐硕士、曹文龙硕士。
有些还未分开却也远在天涯,只能举起酒杯,遥祝前程,感谢远在澳大利亚求学
的张宁宁博士。谢谢你们一直以来对我的真心帮助和支持!
诚然,“共读同窗月,亦结同窗心”,一起经历的点点滴滴,总是在即将分离
时倍感珍贵,感谢同窗好友彭昌彬硕士,感谢室友徐明硕士、茹毅硕士、杨自立
硕士,谢谢你们的三年陪伴以及对我的帮助与支持!
“江山代有人才出,各领风骚数百年”,能够吃苦耐劳、勤奋好学的师弟、
师妹们不断加入到这个大家庭,在我的论文写作期间,给予了我很多的帮助支持!
感谢刘铖硕士硕士、潘金禾硕士、汤梦成硕士、曹闪闪硕士,谢谢你们对我的帮
助与支持!
“见面怜清瘦,呼儿问苦辛”,父亲、母亲疼惜的话语仍在耳边不断响起,
养育恩情,无以回报!父母为了我辛苦了半辈子,岁月再也难掩父母头上长出的
白发,儿且将深深牵挂、深深祝福寄托于字里行间。谢谢你们对我无私的爱!
人生当中终会有一相识、相知、相爱之人陪你走过春夏秋冬,“执子之手,
与子偕老”,这最深情的告白就是我对你最好的感谢!感谢郑宛如女士在我生命
中出现,给予我柔情的爱!
只愿得时光,看看这百年矿大,尝一尝小南门的板面。
最后,衷心感谢在百忙之中评阅本论文的各位专家、学者,由于作者水平有
限,文中难免出现错误和不当之处,敬请不吝赐教。
I
摘 要
在煤泥浮选过程中,浮选泡沫表面视觉特征是浮选工况、工艺指标和生产操
作的直接反映。然而,煤泥浮选操作自动化方面的研究仍处于初期阶段,且主要
依靠人工肉眼观察浮选泡沫状态进而指导生产,这种操作方式具有明显的局限性,
难以保证浮选过程的最优控制。因此,利用最新机器视觉技术,提取浮选泡沫图
像特征参数,开发一套煤泥浮选泡沫图像处理系统,以获取准确有效的特征参数
信息,为现场生产操作提供参考依据。本文的主要研究工作有:
针对煤泥浮选泡沫图像的独特性,基于国内外图像处理技术的研究现状,利
用 MATLAB 软件开发了泡沫图像处理系统。结合选煤厂浮选车间的实际需求,
本系统主要由图像分析、图像增强、图像分割以及特征参数提取等几个模块组成。
研究了浮选泡沫灰度图像以及灰度直方图,提取了图像灰度均值并与煤泥灰
分进行了相关性分析。针对选煤厂浮选车间环境复杂的现状,分别开展了空间域
图像增强和频率域图像增强研究,详细分析对比了各算法的处理效果,在此基础
上提出了一种混合图像增强算法。
研究了经典的图像分割算法,详细分析对比了各个分割算法的优缺点,在此
基础上提出了一种基于改进梯度图像与形态学相结合的分割算法。根据分割的效
果处理图发现,该分割算法能够较好的分割浮选泡沫图像,解决了常见的欠分割
与过分割问题。
研究了能量、惯性矩、熵、相关性、逆差矩五个参数的提取,分别对这些特
征参数和精煤灰分进行了相关性分析。根据这些参数随浮选泡沫状态的变化趋势,
提出纹理复杂度这一具有代表意义的表征参数,发现纹理复杂度波动范围在[80,
145]之间,浮选精煤灰分较好,波动范围在[4.89,7.71]之间。
该论文有图 48 幅,表 2 个,参考文献 84 篇。
关键词:浮选泡沫;图像处理;MATALB;纹理复杂度
II
Abstract
The visual features of surface of flotation bubbles, produced in coal slime flotation
process, directly reflect the performance condition and index of flotation. However,
flotation process is still guided based on the observation of foam state by human in
commercial production, the study on automatic application in coal slime flotation is
still in the preliminary stage. The optimal flotation condition is hard to achieve in
current situation with artificial adjustment. Therefore, developing a coal flotation froth
image processing system, based on the latest machine vision technology, is essential.
With which visual characteristic parameters of flotation froth can be extracted and the
precise and effective feature parameter information can be obtained for further guide of
in commercial production of coal flotation. The following contents are mainly included
in this paper:
Based on the uniqueness of slime flotation froth image and the domestic and
abroad research status of image processing technology, control interface of was firstly
developed for the coal flotation froth image processing system via MATLAB software.
Considering the industrial demand of coal flotation process, the system was consisting
of image analysis module, image enhancement module, image segmentation module,
feature extraction module.
The gray level image and gray histogram of flotation froth image were studied and
the correlation between the mean image gray and ash was also analyzed. According to
the present situation of the complex environment of coal flotation plant, studies on the
spatial domain image enhancement and the frequency domain image enhancement were
carried out, the treatment effect of each algorithm was compared and finally, proposes
a hybrid image enhancement algorithm.
Study on the classic image segmentation algorithm was conducted, and the
advantages and disadvantages of each algorithm were analyzed and compared in detail.
An improved segmentation algorithm was proposed based on the above. The results of
the segmentation indicated that the segmentation algorithm had a better segmentation
flotation froth image, the common problems of under segmentation and over
segmentation were well solved.
The extraction of five parameters, including energy, moment of inertia, entropy,
correlation, and inverse difference moment, were studied and correlation analysis these
III
parameters and ash were conducted as well. According to the variation tendency of
these parameters with the flotation froth state, this paper proposed a parameter of
texture complexity, which has important representative significance. When the
fluctuation range of texture complexity is between 110 and 130, ash content of clean
coal is lower and range from 4.89% to 7.71%.
This paper contains graph 48 pieces, table 2 pieces, and reference 84 pieces.
Keywords:flotation froth; image processing; MATLAB; texture complexity
IV
目 录
摘 要 ............................................................................................................................. I
目 录 .......................................................................................................................... IV
图清单 ..................................................................................................................... VIII
表清单 ........................................................................................................................ XI
1
绪论 ............................................................................................................................ 1
1.1 研究背景与意义..................................................................................................... 1
1.2 课题提出................................................................................................................. 1
1.3 研究内容................................................................................................................. 1
2
文献综述 ................................................................................................................... 3
2.1 矿物浮选泡沫监测系统的研究现状..................................................................... 3
2.2 浮选泡沫图像的预处理研究现状......................................................................... 5
2.3 图像纹理特征提取研究现状................................................................................. 6
2.4 MATLAB 功能介绍 ................................................................................................ 8
2.5 本章小结................................................................................................................. 9
3
灰度图像的图像分析 ............................................................................................. 11
3.1 灰度图像及灰度直方图....................................................................................... 11
3.2 纹理特征提取概述................................................................................................ 13
3.3 纹理参数数据分析............................................................................................... 16
3.4 本章小结............................................................................................................... 19
4
浮选泡沫图像增强的预处理 ................................................................................. 20
4.1 空间域图像增强................................................................................................... 20
4.2 频率域图像增强................................................................................................... 27
4.3 改进的混合图像增强算法................................................................................... 31
4.4 本章小结............................................................................................................... 32
5
图像分割算法 ......................................................................................................... 33
5.1 边缘检测分割算法............................................................................................... 33
5.2 分水岭分割算法................................................................................................... 35
剩余83页未读,继续阅读
资源评论
2301_77550592
- 粉丝: 17
- 资源: 7164
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功