Multivariate image analysis was used to estimate the arsenic concentrations in froths resulting from the flotation of different mixtures of realgar and orpiment particles in a laboratory batch flotation cell. The realgar floated rapidly and in excess of 90% of the mineral could be recovered after 2 minutes, whereas only 48–75% of the orpiment could be recovered in the same time. Textural features, based on grey level co-occurrence matrices (GLCMs), local binary patterns (LBPs), steearable pyramids and textons were used in the analysis. Random forest models could explain approximately 71–77% of the variance in the arsenic using either of the texton, steerable pyramid or LBP features. This was considerably better than what could be obtained with the GLCM features. Monitoring of froth flotation cells was simulated with the batch data. The texton textural features were the most discriminatory with regard to detecting changes in the arsenic content of the froth.
AhonenT, HadidA, PietikäinenM.2006. Face description with local binary patterns: application to face recognition. IEEE Trans Pattern Anal Mach Intell. 28:2037–2041. doi: 10.1109/TPAMI.2006.244
2.
AldrichC, AuretL.2013. Unsupervised process monitoring and fault diagnosis with machine learning methods. London: Springer.
3.
AldrichC, MaraisC, SheanBJ, CilliersJJ.2010. Online monitoring and control of froth flotation systems with machine vision: a review. Int J Miner Process. 96:1–13. doi: 10.1016/j.minpro.2010.04.005
4.
AldrichC, SmithLK, VerrelliDI, BruckhardWJ, KistnerM, AuretL.2014. Multivariate image analysis of a realgar–orpiment batch froth flotation system. Proceedings of the 27th International Minerals Processing Congress (IMPC 2014); Oct 20–24; Chapter 4, 14–22, Santiago, Chile.
5.
AldrichC, UahengoFDL, KistnerM.2015. Particle size estimation in hydrocyclone underflow streams by use of multivariate image analysis. Miner Eng. 70:14–19. doi: 10.1016/j.mineng.2014.08.018
6.
AuretL, AldrichC.2011. Empirical comparison of tree ensemble variable importance measures. Chemometr Intell Lab Syst. 105:157–170. doi: 10.1016/j.chemolab.2010.12.004
7.
AuretL, AldrichC.2012. Interpretation of nonlinear relationships between process variables by use of random forests. Miner Eng. 35:27–42. doi: 10.1016/j.mineng.2012.05.008
8.
BonifaziG, MassacciP, MeloniA.2000. Prediction of complex sulfide flotation performances by a combined 3D fractal and colour analysis of the froths. Miner Eng. 13:737–746. doi: 10.1016/S0892-6875(00)00058-3
9.
BreimanL.2001. Random forests. Mach Learn. 45:5–32. doi: 10.1023/A:1010933404324
10.
CaoB, XieY, GuiW, WeiL, YangC.2013. Integrated prediction model of bauxite concentrate grade based on distributed machine vision. Miner Eng. 53:31–38. doi: 10.1016/j.mineng.2013.07.003
11.
Del VillarR, DesbiensA, MaldonadoM, BouchardJ.2010. Automatic control of flotation columns. In: SbárbaroD, Del VillarR, editor. Advanced control and supervision of mineral processing plants. London: Springer; 249–286.
12.
DoMN, VetterliM.2002. Rotation invariant texture characterization and retrieval using steerable wavelet-domain hidden Markov models. IEEE Trans Multimed. 4:517–527. doi: 10.1109/TMM.2002.802019
13.
DuchesneC.2010. Multivariate image analysis in mineral processing. In: SbárbaroD and Del VillarR, editors. Advanced control and supervision of mineral processing plants. London: Springer; 85–142.
14.
FuY, AldrichC.2016. Flotation froth image analysis by use of a dynamic feature extraction algorithm. Proceedings of the 17th IFAC symposium on control, optimization and automation in mining, mineral and metal processing (MMM 2016);August 31–September 2; Vienna, Austria.
15.
GreenspanH, BelongieS, GoodmanR, PeronaP.1994. Rotation invariant texture recognition using a steerable pyramid. IEEE Proceedings of the 12th IAPR International Conference on Pattern Recognition, October 9–13, Jerusalem, Israel, Vol. 2, p. 162–167.
16.
GuiW, LiuJ, YangC, ChenN, LiaoX.2013. Color co-occurrence matrix based froth image texture extraction for mineral flotation. Miner Eng. 46–47:60–67. doi: 10.1016/j.mineng.2013.03.024
17.
HaqueN, BruckardWJ.Cuevas,J.2012. A techno-economic comparison of pyrometallurgical and hydrometallurgical options for treating high-arsenic copper concentrates. Proceedings of the 26th International Mineral Processing Congress (IMPC 2012); September 2012, New Delhi, India,01908–01922.
18.
HaralickRM, ShanmugamK, DinsteinI.1973. Textural features for image classification. IEEE Trans Syst Man Cybernet. SMC-3:610–621. doi: 10.1109/TSMC.1973.4309314
19.
HeegerDJ, BergenJR.1995. Pyramid-based texture analysis/synthesis. Proceedings of the 22nd Annual Conference on Computer Graphics and Interactive Techniques (SIGGRAPH ‘95), August 6–11, Los Angeles, USA, p. 229–238.
20.
JahedsaravaniA, MarhabanMH, MassinaeiM.2014. Prediction of the metallurgical performances of a batch flotation system by image analysis and neural networks. Miner Eng. 69:137–145. doi: 10.1016/j.mineng.2014.08.003
21.
JemwaGT, AldrichC.2012. Estimating size fraction categories of coal particles on conveyor belts using image texture modeling methods. Expert Syst Appl. 39:7947–7960. doi: 10.1016/j.eswa.2012.01.104
22.
KaartinenJ, HätönenJ, HyötyniemiH, MiettunenJ.2006. Machine-vision-based control of zinc flotation – a case study. Control Eng Pract. 14:1455–1466. doi: 10.1016/j.conengprac.2005.12.004
23.
KistnerM, JemwaGT, AldrichC.2013. Monitoring of mineral processing systems by using textural image analysis. Miner Eng. 52:169–177. doi: 10.1016/j.mineng.2013.05.022
24.
LeungT, MalikJ.2001. Representing and recognizing the visual appearance of materials using three-dimensional textons. Int J Comput Vision. 43:29–44. doi: 10.1023/A:1011126920638
25.
LiS, Shawe-TaylorJ.2005. Comparison and fusion of multiresolution features for texture classification. Pattern Recogn Lett. 26:633–638. doi: 10.1016/j.patrec.2004.09.013
26.
LiuM, TangZ, WangX, GuiW, YangG.2015. Performance recognition of antimony flotation based on multi-information fusion and extension theory. Zhongnan Daxue Xuebao (Ziran Kexue Ban)/J Cent South Univ (Science and Technology). 46:4512–4520.
27.
MaX, BruckardWJ.2009. Rejection of arsenic minerals in sulfide flotation – A literature review. Int J Miner Process. 93:89–94. doi: 10.1016/j.minpro.2009.07.003
28.
MaraisC, AldrichC.2011. Estimation of platinum flotation grades from froth image data. Miner Eng. 24:433–441. doi: 10.1016/j.mineng.2010.12.006
29.
MoolmanDW, AldrichC, Van DeventerJSJ.1995. The monitoring of froth surfaces on industrial flotation plants using connectionist image processing techniques. Miner Eng. 8:23–30. doi: 10.1016/0892-6875(94)00099-X
30.
NanniL, LuminiA, BrahnamS.2012. Survey on LBP based texture descriptors for image classification. Expert Syst Appl. 39:3634–3641. doi: 10.1016/j.eswa.2011.09.054
31.
O'BrienJJ, SpryPG, NettletonD, XuR, TealeGS.2015. Using random forests to distinguish gahnite compositions as an exploration guide to Broken Hill-type Pb–Zn–Ag deposits in the Broken Hill domain, Australia. J Geochem Explor. 149:74–86. doi: 10.1016/j.gexplo.2014.11.010
32.
OjalaT, PietikäinenM, HarwoodD.1994. Performance evaluation of texture measures with classification based on kullback discrimination of distributions. IEEE Proceedings of the 12th IAPR International Conference on Pattern Recognition, October 9–13, Jerusalem, Israel, Vol. 2, p. 582–585.
33.
PortillaJ, SimoncelliEP.2000. A parametric texture model based on joint statistics of complex wavelet coefficients. Int J Mach Vision. 40:49–70. doi: 10.1023/A:1026553619983
34.
Rodriguez-GalianoV, Sanchez-CastilloM, Chica-OlmoM, Chica-RivasM.2015. Machine learning predictive models for mineral prospectivity: an evaluation of neural networks, random forest, regression trees and support vector machines. Ore Geol Rev. 71:804–818. doi: 10.1016/j.oregeorev.2015.01.001
35.
SimoncelliEP, FreemanWT.1995. The steerable pyramid: a flexible architecture for multi-scale derivative computation. IEEE second International Conference on Image Processing; October; Washington DC, USA.
SmithLK, DaveyKJ.BruckardWJ.2012. The use of pulp potential control to separate copper and arsenic – an overview based on selected case studies. Proceedings of the 26th International Mineral Processing Congress (IMPC 2012); September 2012; New Delhi, India, 05057–05067.
38.
SmithLK, DaveyKJ, ForbesL, BruckardWJ.2013. Fundamental studies of the flotation of realgar and orpiment. Proceedings of the XV Balkan Mineral Processing Congress; June 12–16; Sozopol, Bulgaria, 1, 379–384.
39.
TanJ, LiangL, PengY, XieG.2016. The concentrate ash content analysis of coal flotation based on froth images. Miner Eng. 92:9–20. doi: 10.1016/j.mineng.2016.02.006
40.
VarmaM, ZissermanA.2005. A statistical approach to texture classification from single images. Int J Comput Vision. 62:61–81. doi: 10.1007/s11263-005-4635-4
41.
VatsavaiRR, CheriyadatA, GleasonS.2010. Unsupervised semantic labeling framework for identification of complex facilities in high-resolution remote sensing images. IEEE International Workshop on Spatial and Spatiotemporal Data Mining (SSTDM-10), September 13, Sydney, Australia, p. 273–280.
42.
WuJ, XieY-F, YangC-H, GuiW-H.2015. Froth-image-features-driven control strategy for addition rates of reagents in stibnium rougher flotation. Kongzhi Lilun Yu Yingyong/Control Theory and Applications. 32:1599–1606.
43.
ZhangJ, TangZ, LiuJ, TanZ, XuP.2016. Recognition of flotation working conditions through froth image statistical modeling for performance monitoring. Miner Eng. 86:116–129. doi: 10.1016/j.mineng.2015.12.008
