Coal-rock image recognition method integrating drilling geological information
-
摘要: 当前应用于煤岩图像识别的深度卷积神经网络模型存在体积庞大、计算过程冗杂等问题，难以满足实时检测要求，且对低照度、高粉尘等复杂环境适应性差。针对上述问题，提出了一种融合钻孔地质信息的煤岩图像识别方法。首先，通过改进的谱残差显著性检测（ISRSD）算法增强煤岩图像质量，有效减弱复杂环境对煤岩图像特征造成的不利影响；然后，使用加入注意力机制的VGG（AVGG）深度卷积神经网络模型——在VGG的基础上进行剪枝、加入卷积注意力模块（CBAM）和引入自适应学习率调整策略，高效提取煤岩图像特征；最后，利用贝叶斯模型融合煤岩图像特征和由钻孔地质柱状图获取的钻孔地质信息，提升煤岩分类的准确性和鲁棒性。实验结果表明，经ISRSD算法增强后的图像目标更突出，色彩失真程度更低，且边缘、纹理等图像特征保留相对完整； AVGG模型的准确率与VGG模型相当，但平均推理时间、参数量及模型大小分别仅为VGG模型的15.61%，33.44%及33.40%；与仅使用AVGG模型识别煤岩图像相比，利用贝叶斯模型融合钻孔地质信息后，准确率提高了1.85%，达97.31%。Abstract: The current deep convolutional neural network models applied to coal-rock image recognition have problems such as large volume and cumbersome calculation process. It is difficult to meet real-time detection requirements, and it has poor adaptability to complex environments such as low lighting and high dust. In order to solve the above problems, a coal-rock image recognition method integrating drilling geological information is proposed. Firstly, the improved spectral residual saliency detection (ISRSD) algorithm is used to enhance the quality of coal-rock images, effectively reducing the adverse effects of complex environments on the features of coal-rock images. Secondly, the method uses the attentional VGG (AVGG) deep convolutional neural network model. The AVGG performs pruning based on VGG, adds convolutional block attention module (CBAM), and introduces adaptive learning rate adjustment strategy to efficiently extract coal-rock image features. Finally, the Bayesian model is used to integrate the features of coal-rock images with the geological information obtained from the borehole geological column chart, in order to improve the accuracy and robustness of coal-rock classification. The experimental results show that the image enhanced by the ISRSD algorithm has more prominent targets, lower color distortion, and relatively complete preservation of image features such as edges and textures. The accuracy of the AVGG model is comparable to that of the VGG model, but the average inference time, parameter count, and model size are only 15.61%, 33.44%, and 33.40% of the VGG model, respectively. Compared with using only the AVGG model to recognize coal-rock images, using the Bayesian model to fuse drilling geological information improves accuracy by 1.85%, reaching 97.31%.
-
图 1 融合钻孔地质信息的煤岩图像识别方法流程
Figure 1. Process of coal-rock image recognition method integrating drilling geological information
图 2 ISRSD算法流程
Figure 2. Improved spectral residual saliency detection（ISRSD） algorithm flow
图 6 不同显著性检测算法增强效果对比
Figure 6. Comparison of enhancement effects of different saliency detection algorithms
图 7 不同显著性检测算法下AVGG模型的准确率
Figure 7. Accuracy of attentional visual geometry group（AVGG） model under different saliency detection algorithms
图 8 融合钻孔地质信息前后煤岩识别准确率对比
Figure 8. Comparison of coal-rock recognition accuracy before and after fusion borehole geological information
表 1 数据集扩充及类别均衡前后样本数量
Table 1. Sample size before and after data set augmentation and category balancing
类别 数据集扩充及类别
均衡前样本数/张数据集扩充及类别
均衡后样本数/张砂岩 337 1 515 砾岩 96 1 344 泥岩 369 1 348 煤炭 360 1 440 页岩 82 1 394 
下载: 导出CSV
表 2 不同模型评价指标结果对比
Table 2. Comparison of evaluation index results of different models
模型 准确率/% 单个样本
平均推理时间/ms参数量/106个 模型大小
/MiBGoogLeNet 89.65 68.49 5.981 24 VGG 93.79 117.65 128.810 515 SVGG 83.72 18.86 43.081 172 AVGG 91.33 18.36 43.084 172
下载: 导出CSV
表 3 不同显著性检测算法耗时对比
Table 3. Comparison of time consumption of different saliency detection algorithms
算法 单个样本耗时/ms ISRSD 7.9 FT 5.5 LC 15.9 HC 194.3
下载: 导出CSV
-
[1] 王国法,刘峰,庞义辉,等. 煤矿智能化——煤炭工业高质量发展的核心技术支撑[J]. 煤炭学报,2019,44(2):349-357.WANG Guofa,LIU Feng,PANG Yihui,et al. Coal mine intellectualization:the core technology of high quality development[J]. Journal of China Coal Society,2019,44(2):349-357. [2] 王伟,李擎,张德政,等. 基于深度学习的矿石图像处理研究综述[J]. 工程科学学报,2023,45(4):621-631.WANG Wei,LI Qing,ZHANG Dezheng,et al. A survey of ore image processing based on deep learning[J]. Chinese Journal of Engineering,2023,45(4):621-631. [3] 杨健健,张强,王超,等. 煤矿掘进机的机器人化研究现状与发展[J]. 煤炭学报,2020,45(8):2995-3005.YANG Jianjian,ZHANG Qiang,WANG Chao,et al. Status and development of robotization research on roadheader for coal mines[J]. Journal of China Coal Society,2020,45(8):2995-3005. [4] 陈浜. 基于视觉计算的煤岩识别方法研究[D]. 北京:中国矿业大学(北京),2018.CHEN Bang. Methodological studies of coal-rock recognition through visual computing[D]. Beijing:China University of Mining & Technology-Beijing,2018. [5] 张云,童亮,来兴平,等. 基于机器视觉的煤尘环境下掘进空间煤岩界面感知与精准识别[J]. 煤炭学报,2024,49(7):3276-3290.ZHANG Yun,TONG Liang,LAI Xingping,et al. Coal-rock interface perception and accurate recognition in heading face under coal dust environment based on machine vision[J]. Journal of China Coal Society,2024,49(7):3276-3290. [6] 王建才,李瑾,李志军,等. 基于改进 YOLOv5 的煤岩图像检测识别研究[J]. 煤矿机械,2022,43(9):204-208.WANG Jiancai,LI Jin,LI Zhijun,et al. Research on coal rock image detection and recognition based on improved YOLOv5[J]. Coal Mine Machinery,2022,43(9):204-208. [7] 高峰,殷欣,刘泉声,等. 基于塔式池化架构的采掘工作面煤岩图像识别方法[J]. 煤炭学报,2021,46(12):4088-4102.GAO Feng,YIN Xin,LIU Quansheng,et al. Coal-rock image recognition method for mining and heading face based on spatial pyramid pooling structure[J]. Journal of China Coal Society,2021,46(12):4088-4102. [8] 司垒,王忠宾,熊祥祥,等. 基于改进 U−net 网络模型的综采工作面煤岩识别方法[J]. 煤炭学报,2021,46(增刊1):578-589.SI Lei,WANG Zhongbin,XIONG Xiangxiang,et al. Coal-rock recognition method of fully-mechanized coal mining face based on improved U-net network model[J]. Journal of China Coal Society,2021,46(S1):578-589. [9] 张斌,苏学贵,段振雄,等. YOLOv2在煤岩智能识别与定位中的应用研究[J]. 采矿与岩层控制工程学报,2020,2(2):90-97.ZHANG Bin,SU Xuegui,DUAN Zhenxiong,et al. Application of YOLOv2 in intelligent recognition and location of coal and rock[J]. Journal of Mining and Strata Control Engineering,2020,2(2):90-97. [10] 闫志蕊,王宏伟,耿毅德. 基于改进DeeplabV3+和迁移学习的煤岩界面图像识别方法[J]. 煤炭科学技术,2023,51(增刊1):429-439.YAN Zhirui,WANG Hongwei,GENG Yide. Coal-rock interface image recognition method based on improved DeeplabV3+ and transfer learning[J]. Coal Science and Technology,2023,51(S1):429-439. [11] 伍云霞,孟祥龙. 局部约束的自学习煤岩识别方法[J]. 煤炭学报,2018,43(9):2639-2646.WU Yunxia,MENG Xianglong. Locality-constrained self-taught learning for coal-rock recognition[J]. Journal of China Coal Society,2018,43(9):2639-2646. [12] 伍云霞,田一民. 基于字典学习的煤岩图像特征提取与识别方法[J]. 煤炭学报,2016,41(12):3190-3196.WU Yunxia,TIAN Yimin. Method of coal-rock image feature extraction and recognition based on dictionary learning[J]. Journal of China Coal Society,2016,41(12):3190-3196. [13] 吴岚虎,李智玮,刘垒烨,等. 基于场景几何信息的显著性目标检测方法综述[J]. 模式识别与人工智能,2023,36(2):120-142.WU Lanhu,LI Zhiwei,LIU Leiye,et al. A survey of salient object detection based on scene geometric information[J]. Pattern Recognition and Artificial Intelligence,2023,36(2):120-142. [14] HOU Xiaodi,ZHANG Liqing. Saliency detection:a spectral residual approach[C]. IEEE Conference on Computer Vision and Pattern Recognition,Minneapolis,2007:1-8. [15] SIMONYAN K,ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[J]. Computer Science, 2014. DOI: 10.48550/arXiv.1409.1556. [16] WOO S,PARK J,LEE J Y,et al. CBAM:convolutional block attention module[C]. European Conference on Computer Vision,Munich,2018:61-78. [17] HARRINGTON P. Machine learning in action[M]. Singapore:Springer Nature Singapore Pte Ltd. ,2012. [18] ZHOU Zhihua. Machine learning[M]. Singapore:Springer Nature Singapore Pte Ltd. ,2021. [19] ACHANTA R,HEMAMI S,ESTRADA F,et al. Frequency-tuned salient region detection[C]. IEEE Conference on Computer Vision and Pattern Recognition,Munich ,2009:1597-1604. [20] CHENG Mingming,ZHANG Guoxin,MITRA N J,et al. Global contrast based salient region detection[C]. IEEE Transactions on Pattern Analysis and Machine Intelligence,Santa Barbara,2015:569-582. [21] ZHAI Yun,SHAH M. Visual attention detection in video sequences using spatiotemporal cues[C]. The 14th ACM International Conference on Multimedia,Santa Barbara,2006:815-824. -
点击查看大图
计量
- 文章访问数: 76
- HTML全文浏览量: 18
- PDF下载量: 3
- 被引次数: 0
