摘要
针对现有煤与矸石图像处理方法存在提取特征参数少、识别精度低等问题,提出了一种融合局部二值模式和灰度共生矩阵的煤与矸石图像纹理特征提取方法。首先将煤与矸石预处理后的图像转换为局部二值模式图像,再利用该图像生成灰度共生矩阵,以角二阶距、相关性、对比度和熵作为纹理特征进行均值和归一化处理,最后用支持向量机进行训练,得出识别结果。实验结果表明,该方法能够有效地提取到煤与矸石图像的纹理特征,煤和矸石的识别率分别为94%和96%。
In view of problems of less extraction feature parameters and low recognition precision existed in image processing methods of coal and gangue,an extraction method of texture feature of images of coal and gangue fused with local binary pattern and gray level co-occurrence matrix was proposed.Firstly,the preprocessed images of coal and gangue were transformed into local binary pattern images,then the local binary pattern images were used to generate gray level co-occurrence matrix,the mean value and normalization of those texture features including angular second moment,correlation,contrast and entropy were processed.Finally,support vector machine was used for samples training and recognition results were obtained.The experimental results show that the method can effectively extract the texture feature of images of coal and gangue,and the recognition rates of coal and gangue are respectively 94%and96%.
出处
《工矿自动化》
北大核心
2017年第5期26-30,共5页
Journal Of Mine Automation
基金
山东省研究生教育创新计划项目(01040105305)
山东科技大学教学研究项目(JG201506)
山东科技大学研究生教育创新项目(KDYC13026
KDYC15019)
关键词
煤与矸石
图像处理
纹理特征
局部二值模式
灰度共生矩阵
支持向量机
coal and gangue
image processing
texture feature
local binary pattern
gray level co-occurrence matrix
support vector machine