浮选过程中的一种基于改进的 U-Net++ 语义分割的泡沫速度测量方法

陈一玮; 徐德刚; 万坤

doi:10.1007/s12613-023-2787-2

Volume 31 Issue 8

Aug. 2024

Turn off MathJax

图(13) / 表(2)

数据统计

计量

文章访问数: 413
HTML全文浏览量: 195
PDF下载量: 40
被引次数: 0

文章导航 > 矿物冶金与材料学报（英文） > 2024 > 31(8): 1816-1827

Yiwei Chen, Degang Xu, and Kun Wan, A froth velocity measurement method based on improved U-Net++ semantic segmentation in flotation process, Int. J. Miner. Metall. Mater., 31(2024), No. 8, pp. 1816-1827. https://doi.org/10.1007/s12613-023-2787-2

Cite this article as:

Yiwei Chen, Degang Xu, and Kun Wan, A froth velocity measurement method based on improved U-Net++ semantic segmentation in flotation process, Int. J. Miner. Metall. Mater., 31(2024), No. 8, pp. 1816-1827. https://doi.org/10.1007/s12613-023-2787-2

Citation:

PDF( 4592 KB)

引用本文 PDF XML SpringerLink

研究论文

浮选过程中的一种基于改进的 U-Net++ 语义分割的泡沫速度测量方法

中南大学自动化学院, 长沙 410083, 中国

通讯作者:
徐德刚 E-mail: dgxu@csu.edu.cn

文章亮点

(1) 手动构建了钨精矿浮选图像数据集。

(2) 利用形变场对图片及标签自适应地生成变换，以达到数据增强的目的。

(3) 提出了一种适用于浮选泡沫图像的改进分割网络，提升了分割结果的精度和准确性。

(4) 利用语义分割的结果，实现了泡沫实时流速的提取。

中文摘要

在浮选过程中，泡沫图像的特征参数和精矿品位和对应工况高度相关。在不同的工况下，气泡的颜色、大小等静态特征和速度等动态特征有着明显的区别。这些特征的提取通常依赖于泡沫边缘的图像分割结果，因此泡沫图像的分割是研究其视觉信息的基础。同时，由于缺乏科学可靠的带标签的训练数据，且必须手动构建数据集和标签，这给矿物浮选研究带来了困难。为解决这一问题，本文构建了钨精矿浮选图像数据集，并提出了基于cGAN 的数据增强网络和基于 U-Net++ 的泡沫图像分割网络。本文测试了该算法的性能，并与其他算法进行了比较。最后，还基于语义分割的结果，提出了基于相位相关的泡沫流速特征检测方法。

关键词:

Research Article

A froth velocity measurement method based on improved U-Net++ semantic segmentation in flotation process

+ Author Affiliations

Abstract

Abstract

During flotation, the features of the froth image are highly correlated with the concentrate grade and the corresponding working conditions. The static features such as color and size of the bubbles and the dynamic features such as velocity have obvious differences between different working conditions. The extraction of these features is typically relied on the outcomes of image segmentation at the froth edge, making the segmentation of froth image the basis for studying its visual information. Meanwhile, the absence of scientifically reliable training data with label and the necessity to manually construct dataset and label make the study difficult in the mineral flotation. To solve this problem, this paper constructs a tungsten concentrate froth image dataset, and proposes a data augmentation network based on Conditional Generative Adversarial Nets (cGAN) and a U-Net++-based edge segmentation network. The performance of this algorithm is also evaluated and contrasted with other algorithms in this paper. On the results of semantic segmentation, a phase-correlation-based velocity extraction method is finally suggested.
- froth flotation;
- froth segmentation;
- froth image;
- data augmentation;
- velocity extraction;
- image features

FullText(HTML)

Supplements(0)

References(43)

References

[1]	Y.L. Lu, D.W. Liu, X.D. Jia, J.J. Yuan, and D.Y. Shi, A review on flotation process of scheelite, Adv. Mater. Res., 962-965(2014), p. 388. doi: 10.4028/www.scientific.net/AMR.962-965.388
[2]	Z.Y. Chang, S.S. Niu, Z.C. Shen, L.C. Zou, and H.J. Wang, Latest advances and progress in the microbubble flotation of fine minerals: Microbubble preparation, equipment, and applications, Int. J. Miner. Metall. Mater., 30(2023), No. 7, p. 1244. doi: 10.1007/s12613-023-2615-8
[3]	D.W. Moolman, C. Aldrich, J.S.J.Van Deventer, and W.W. Stange, Digital image processing as a tool for on-line monitoring of froth in flotation plants, Miner. Eng., 7(1994), No. 9, p. 1149. doi: 10.1016/0892-6875(94)00058-1
[4]	D.W. Moolman, J.J. Eksteen, C. Aldrich, and J.S.J. van Deventer, The significance of flotation froth appearance for machine vision control, Int. J. Miner. Process., 48(1996), No. 3-4, p. 135. doi: 10.1016/S0301-7516(96)00022-1
[5]	D.W. Moolman, C. Aldrich, J.S.J. Van Deventer, and D.J. Bradshaw, The interpretation of flotation froth surfaces by using digital image analysis and neural networks, Chem. Eng. Sci., 50(1995), No. 22, p. 3501. doi: 10.1016/0009-2509(95)00190-G
[6]	W. Wang, F. Bergholm, and B. Yang, Froth delineation based on image classification, Miner. Eng., 16(2003), No. 11, p. 1183. doi: 10.1016/j.mineng.2003.07.014
[7]	W.X. Wang and O. Stephansson, A robust bubble delineation algorithm for froth images, [in] Proceedings of the Second International Conference on Intelligent Processing and Manufacturing of Materials. IPMM'99, Honolulu, 2002, p. 471.
[8]	A. Jahedsaravani, M. Massinaei, and M.H. Marhaban, An image segmentation algorithm for measurement of flotation froth bubble size distributions, Measurement, 111(2017), p. 29. doi: 10.1016/j.measurement.2017.07.023
[9]	J. Zhang, Z.H. Tang, M.X. Ai, and W.H. Gui, Nonlinear modeling of the relationship between reagent dosage and flotation froth surface image by Hammerstein-Wiener model, Miner. Eng., 120(2018), p. 19. doi: 10.1016/j.mineng.2018.01.018
[10]	J.M. Hargrave and S.T. Hall, Diagnosis of concentrate grade and mass flowrate in tin flotation from colour and surface texture analysis, Miner. Eng., 10(1997), No. 6, p. 613. doi: 10.1016/S0892-6875(97)00040-X
[11]	C. Marais and C. Aldrich, Estimation of platinum flotation grades from froth image data, Miner. Eng., 24(2011), No. 5, p. 433. doi: 10.1016/j.mineng.2010.12.006
[12]	K. Popli, A. Afacan, Q. Liu, and V. Prasad, Development of online soft sensors and dynamic fundamental model-based process monitoring for complex sulfide ore flotation, Miner. Eng., 124(2018), p. 10. doi: 10.1016/j.mineng.2018.04.006
[13]	J. Zhang, Z.H. Tang, Y.F. Xie, M.X. Ai, and W.H. Gui, Convolutional memory network-based flotation performance monitoring, Miner. Eng., 151(2020), art. No. 106332. doi: 10.1016/j.mineng.2020.106332
[14]	Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner, Gradient-based learning applied to document recognition, Proc. IEEE, 86(1998), No. 11, p. 2278. doi: 10.1109/5.726791
[15]	A. Krizhevsky, I. Sutskever, and G.E. Hinton, ImageNet classification with deep convolutional neural networks, Commun. ACM, 60(2017), No. 6, p. 84. doi: 10.1145/3065386
[16]	K. Simonyan and A. Zisserman, Very deep convolutional networks for large-scale image recognition, [in] International Conference on Learning Representations, San Diego, 2015.
[17]	K.M. He, X.Y. Zhang, S.Q. Ren, and J. Sun, Deep residual learning for image recognition, [in] 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR ), Las Vegas, 2016, p. 770.
[18]	H. Noh, S. Hong, and B. Han, Learning deconvolution network for semantic segmentation, [in] 2015 IEEE International Conference on Computer Vision (ICCV ), Santiago, 2015, p. 1520.
[19]	W.G. Baxt, Use of an artificial neural network for the diagnosis of myocardial infarction, Ann. Intern. Med., 115(1991), No. 11, p. 843. doi: 10.7326/0003-4819-115-11-843
[20]	R. Girshick, J. Donahue, T. Darrell, and J. Malik, Rich feature hierarchies for accurate object detection and semantic segmentation, [in] 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, 2014, p. 580.
[21]	D. Forsyth, Object detection with discriminatively trained part-based models, Computer, 47(2014), No. 2, p. 6. doi: 10.1109/MC.2014.42
[22]	J. Wang, Y. Yang, J.H. Mao, Z.H. Huang, C. Huang, and W. Xu, CNN-RNN: A unified framework for multi-label image classification, [in] 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR ), Las Vegas, 2016, p. 2285.
[23]	A. Garcia-Garcia, S. Orts-Escolano, S.O. Oprea, V. Villena-Martinez, and J. Garcia-Rodriguez, A review on deep learning techniques applied to semantic segmentation, 2017. https://arxiv.org/abs/1704.06857v1.
[24]	J. Long, E. Shelhamer, and T. Darrell, Fully convolutional networks for semantic segmentation, [in] 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR ), Boston, 2015, p. 3431.
[25]	L.C. Chen, G. Papandreou, I. Kokkinos, K. Murphy, and A.L. Yuille, DeepLab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs, IEEE Trans. Pattern Anal. Mach. Intell., 40(2018), No. 4, p. 834. doi: 10.1109/TPAMI.2017.2699184
[26]	J.P. Liu, Q.Q. Gao, Z.H. Tang, et al., Online monitoring of flotation froth bubble-size distributions via multiscale deblurring and multistage jumping feature-fused full convolutional networks, IEEE Trans. Instrum. Meas., 69(2020), No. 12, p. 9618. doi: 10.1109/TIM.2020.3006629
[27]	B.K. Gharehchobogh, Z.D. Kuzekanani, J. Sobhi, and A.M. Khiavi, Flotation froth image segmentation using Mask R-CNN, Miner. Eng., 192(2023), art. No. 107959. doi: 10.1016/j.mineng.2022.107959
[28]	O. Ronneberger, P. Fischer, and T. Brox, U-Net: Convolutional networks for biomedical image segmentation, [in] N. Navab, J. Hornegger, W.M. Wells, and AF. Frangi, eds., Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, Part III, Munich, 2015, p. 234.
[29]	Z.W. Zhou, M.M.R. Siddiquee, N. Tajbakhsh, and J.M. Liang, UNet++: A nested U-Net architecture for medical image segmentation, [in] D. Stoyanov, Z. Taylor, G. Carneiro, et al., eds., Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support (DLMIA 2018 , ML-CDS 2018 ), Granada, 2018, p. 3.
[30]	A. Dosovitskiy, P. Fischer, J.T. Springenberg, M. Riedmiller, and T. Brox, Discriminative unsupervised feature learning with exemplar convolutional neural networks, IEEE Transactions on Pattern Analysis and Machine Intelligence, 38(2016), No. 9, p. 1734. doi: 10.1109/TPAMI.2015.2496141
[31]	Z.X. Jiang, H. Zhang, Y. Wang, and S.B. Ko, Retinal blood vessel segmentation using fully convolutional network with transfer learning, Comput. Med. Imag. Graph., 68(2018), p. 1. doi: 10.1016/j.compmedimag.2018.04.005
[32]	I.J. Goodfellow, J. Pouget-Abadie, M. Mirza, et al., Generative adversarial networks, Commun. ACM, 63(2020), No. 11, p. 139. doi: 10.1145/3422622
[33]	X. Yi, E. Walia, and P. Babyn, Generative adversarial network in medical imaging: A review, Med. Image Anal., 58(2019), art. No. 101552. doi: 10.1016/j.media.2019.101552
[34]	M. Mirza and S. Osindero, Conditional generative adversarial nets, 2014. https://arxiv.org/abs/1411.1784
[35]	A. Radford, L. Metz, and S. Chintala, Unsupervised representation learning with deep convolutional generative adversarial networks, 2015. https://arxiv.org/abs/1511.06434
[36]	C. Szegedy, W. Liu, Y.Q. Jia, et al., Going deeper with convolutions, [in] 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR ), Boston, 2015, p. 1.
[37]	N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, Dropout: A simple way to prevent neural networks from overfitting, J. Mach. Learn. Res., 15(2014), No. 1, p. 1929.
[38]	A. Painsky and G. Wornell, On the universality of the logistic loss function, [in] 2018 IEEE International Symposium on Information Theory (ISIT ), Vail, 2018, p. 936.
[39]	P. Ramachandran, B. Zoph, and Q.V. Le, Searching for activation functions, 2017. http://arxiv.org/abs/1710.05941
[40]	J.M. Hargrave, N.J. Miles, and S.T. Hall, The use of grey level measurement in predicting coal flotation performance, Miner. Eng., 9(1996), No. 6, p. 667. doi: 10.1016/0892-6875(96)00054-4
[41]	A. Jahedsaravani, M.H. Marhaban, and M. Massinaei, Prediction of the metallurgical performances of a batch flotation system by image analysis and neural networks, Miner. Eng., 69(2014), p. 137. doi: 10.1016/j.mineng.2014.08.003
[42]	M. Massinaei, A. Jahedsaravani, E. Taheri, and J. Khalilpour, Machine vision based monitoring and analysis of a coal column flotation circuit, Powder Technol., 343(2019), p. 330. doi: 10.1016/j.powtec.2018.11.056
[43]	Y.L. Zhou and H.W. Li, The analysis of gas-liquid two-phase flow patterns based on variation coefficient of image connected regions and line-correlation algorithm, Energy Procedia, 17(2012), p. 933. doi: 10.1016/j.egypro.2012.02.190