Academia.eduAcademia.edu

Figure 7 – uploaded by Roberto Galery

See full PDFdownloadDownload figure
Fig. 7. (a) Pre-processed digital image from RLOM; (b) segmented image (clustering and sub-clustering); (c) image after “UnsharpMask’” filter application; (d) image (c) after segmentation resulting in the enhancing of the borders, pores and cracks; (e) image produced by addition of images (b) and (d).

Figure 7 (a) Pre-processed digital image from RLOM; (b) segmented image (clustering and sub-clustering); (c) image after “UnsharpMask’” filter application; (d) image (c) after segmentation resulting in the enhancing of the borders, pores and cracks; (e) image produced by addition of images (b) and (d).

Related Figures (12)

Fig. 1. Images sampled from the same area by different systems: (a) SEM; (b) RLOM.
Fig. 1. Images sampled from the same area by different systems: (a) SEM; (b) RLOM.
Polishing parameters. Table 1
Polishing parameters. Table 1
Fig. 2. Particle images and respective image histograms taken from different areas of the same polished section for the 150-106-1m size class
Fig. 2. Particle images and respective image histograms taken from different areas of the same polished section for the 150-106-1m size class
Fig. 3. RLOM image taken from iron ore sample containing quartz (Qz), goethite (Gt), magnetite (Mt), and hematite (Ht).
Fig. 3. RLOM image taken from iron ore sample containing quartz (Qz), goethite (Gt), magnetite (Mt), and hematite (Ht).
‘ig. 4. (a) Graphical representation of the N x M mask recursively applied for boundary definition; (b) RLOM image with sub-clusters exhibiting defined borders, pores, an racks; (c) image (b) with pseudo-colours attributed to each of the clusters: borders, pores, and cracks (white), quartz/resin (red), goethite (blue), magnetite (green), an 1ematite (magenta). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) by the s(G) seed points (in green). From left to right, G, C3, and C, represent, respectively, goethite, magnetite, and hematite; the three “homogeneous” clusters on the grey-level scale. (OPS) Find a partition, Py € zy, that provides the most homoge- nous (discriminative) classification of the set of grey-level intensities, I, as expressed in terms of the dissimilarities, D.
‘ig. 4. (a) Graphical representation of the N x M mask recursively applied for boundary definition; (b) RLOM image with sub-clusters exhibiting defined borders, pores, an racks; (c) image (b) with pseudo-colours attributed to each of the clusters: borders, pores, and cracks (white), quartz/resin (red), goethite (blue), magnetite (green), an 1ematite (magenta). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) by the s(G) seed points (in green). From left to right, G, C3, and C, represent, respectively, goethite, magnetite, and hematite; the three “homogeneous” clusters on the grey-level scale. (OPS) Find a partition, Py € zy, that provides the most homoge- nous (discriminative) classification of the set of grey-level intensities, I, as expressed in terms of the dissimilarities, D.
Fig. 6. Selected section of Fig. 5b showing quartz border discontinuities (arrows) after the segmentation process. Fig. 5. (a) RLOM digital image from an iron ore sample showing the mineral phases containing quartz (Qz), goethite (Gt), magnetite (Mt), and hematite (Ht); (b) image (a) after the segmentation procedure described in Sections 3.2 and 3.3.
Fig. 6. Selected section of Fig. 5b showing quartz border discontinuities (arrows) after the segmentation process. Fig. 5. (a) RLOM digital image from an iron ore sample showing the mineral phases containing quartz (Qz), goethite (Gt), magnetite (Mt), and hematite (Ht); (b) image (a) after the segmentation procedure described in Sections 3.2 and 3.3.
Fig. 8. (a) Image with quartz border reinforcement; (b) segmented input image (binarisation); (c) filling process applied to image (b); (d) image (c) following erosion-dilation process; (e) images (a) + (c); (f) logic operation applied to images (d) and (e) to fill and enhance particle pores and cracks.
Fig. 8. (a) Image with quartz border reinforcement; (b) segmented input image (binarisation); (c) filling process applied to image (b); (d) image (c) following erosion-dilation process; (e) images (a) + (c); (f) logic operation applied to images (d) and (e) to fill and enhance particle pores and cracks.
Mineral phase quantification performed using the Opt_,;, and MLA for the size classes: 300-212, 212-150, and 150-106 um. Table 2
Mineral phase quantification performed using the Opt_,;, and MLA for the size classes: 300-212, 212-150, and 150-106 um. Table 2
Fig. 9. Opt.;, and MLA mineral phase quantification for the size classes: 300-212, 212-150, and 150-106 tum.
Fig. 9. Opt.;, and MLA mineral phase quantification for the size classes: 300-212, 212-150, and 150-106 tum.
Fig. 10. Cumulative particle composition distribution obtained from MLA and Opt-:i, for the size classes: 300-212, 212-150, and 150-106 jim.
Fig. 10. Cumulative particle composition distribution obtained from MLA and Opt-:i, for the size classes: 300-212, 212-150, and 150-106 jim.
Related topics:
Chemical EngineeringMaterials Science
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved