12
of values, indicating that the GLCM features for BC are
tightly clustered and the textures tend to be more consis-
tent than that of BBC.
In addition, the influence of resolution on the calcu-
lated GLCM features was investigated. The resolution can
affect the level of details that an image can capture and
thus vary the image textures. The results show that if the
image features are extracted from the patches of the same
size (area), the increase in resolution can help capture finer
details without losing the global texture information. It
helps to capture consistent image features with less varia-
tion, leading to the narrower distribution of the features.
Thus, the increase in resolution can help capture the differ-
ence in texture between different coal lithotypes.
Finally, the influence of camera settings on the image
features was studied. Different settings with shutter speed,
F-stop, and ISO were combined to take rib photos and
varying exposures can be observed in the photos through
visual inspection. The calculated image features show varia-
tion with the exposure. The images with lower luminance
have lower contrast, dissimilarity, deviation, and mean val-
ues and higher energy and homogeneity values, while the
images with higher luminance have higher mean value. The
image preprocessing step with normalization and histogram
equalization can effectively reduce the variations resulted
from camera settings. However, it increases the difficulty
for lithotype classification.
LIMITATIONS
The work completed in this study was from an exploratory
research perspective to evaluate the effectiveness of using
GLCM method to classify coal lithotypes of photos previ-
ously taken by NIOSH researchers at field sites. The find-
ings are limited to the specific lithologies analyzed during
the study from the photos available for a proof-of-concept
evaluation and limited to the camera (Nikon D550) and
shutter speed, F-stop, ISO, and other default settings used.
The application of the proposed method would need to be
validated for other cameras, settings, mine sites, and mine
conditions for broader concluding remarks on the effec-
tiveness of using this technique for automated lithological
detection.
DISCLAIMER
The findings and conclusions in this study are those of the
authors and do not necessarily represent the official posi-
tion of the National Institute for Occupational Safety
and Health (NIOSH), Centers for Disease Control and
Prevention (CDC). Mention of any company or product
does not constitute endorsement by NIOSH.
REFERENCES
[1] Mohamed KM, Van Dyke M, Rashed G, et al (2020)
Preliminary rib support requirements for solid coal
ribs using a coal pillar rib rating. In: Proceeding of the
39th International Conference on Ground Control in
Mining. Canonsburg, PA, USA, pp 85–96.
[2] Rashed G, Barton T, Sears M, et al (2018) Estimation
of the intact strength of coal using indirect methods.
In: Proceedings of the 37th International Conference
on Ground Control in Mining. Morgantown, WV,
USA, pp 294–301.
[3] Rusnak JA (2017) Coal strength variation by lithot-
ype for high-volatile A bituminous coal in the cen-
tral Appalachian Basin. In: Proceedings of the 36th
International Conference on Ground Control in
Mining. Morgantown, WV USA, pp 198–207.
[4] Singh A, Armstrong RT, Regenauer-Lieb K,
Mostaghimi P (2019) Rock Characterization Using
Gray-Level Co-Occurrence Matrix: An Objective
Perspective of Digital Rock Statistics. Water Resour Res
55:1912–1927. doi.org/10.1029/2018WR023342.
[5] Slaker BA, Mohamed KM (2016) A practical appli-
cation of photogrammetry to performing rib char-
acterization measurements in an underground coal
mine using a DSLR camera. In: Proceeding of the
35th International Conference on Ground Control
in Mining. pp 179–186.
[6] Stopes MC (1919) On the four visible ingredients in
banded bituminous coal. In: Proceedings of the Royal
Society of London. Series B, Containing Papers of a
Biological Character. pp 470–487.
[7] Sun J, Su B (2013) Coal–rock interface detection
on the basis of image texture features. Int J Min Sci
Technol 23:681–687.
[8] Xue Y (2022) Coal and Rock Classification with
Rib Images and Machine Learning Techniques.
Mining, Metall Explor 39:453–465. doi.org/10.1007
/s42461-021-00526-4.
[9] Yu K, Barry JC, Esterle JS (1997) Analysis of coal
banding texture and implications for megascopic
image analysis. Int J Coal Geol 33:1–18. doi.org
/10.1016/S0166-5162(96)00022-5
of values, indicating that the GLCM features for BC are
tightly clustered and the textures tend to be more consis-
tent than that of BBC.
In addition, the influence of resolution on the calcu-
lated GLCM features was investigated. The resolution can
affect the level of details that an image can capture and
thus vary the image textures. The results show that if the
image features are extracted from the patches of the same
size (area), the increase in resolution can help capture finer
details without losing the global texture information. It
helps to capture consistent image features with less varia-
tion, leading to the narrower distribution of the features.
Thus, the increase in resolution can help capture the differ-
ence in texture between different coal lithotypes.
Finally, the influence of camera settings on the image
features was studied. Different settings with shutter speed,
F-stop, and ISO were combined to take rib photos and
varying exposures can be observed in the photos through
visual inspection. The calculated image features show varia-
tion with the exposure. The images with lower luminance
have lower contrast, dissimilarity, deviation, and mean val-
ues and higher energy and homogeneity values, while the
images with higher luminance have higher mean value. The
image preprocessing step with normalization and histogram
equalization can effectively reduce the variations resulted
from camera settings. However, it increases the difficulty
for lithotype classification.
LIMITATIONS
The work completed in this study was from an exploratory
research perspective to evaluate the effectiveness of using
GLCM method to classify coal lithotypes of photos previ-
ously taken by NIOSH researchers at field sites. The find-
ings are limited to the specific lithologies analyzed during
the study from the photos available for a proof-of-concept
evaluation and limited to the camera (Nikon D550) and
shutter speed, F-stop, ISO, and other default settings used.
The application of the proposed method would need to be
validated for other cameras, settings, mine sites, and mine
conditions for broader concluding remarks on the effec-
tiveness of using this technique for automated lithological
detection.
DISCLAIMER
The findings and conclusions in this study are those of the
authors and do not necessarily represent the official posi-
tion of the National Institute for Occupational Safety
and Health (NIOSH), Centers for Disease Control and
Prevention (CDC). Mention of any company or product
does not constitute endorsement by NIOSH.
REFERENCES
[1] Mohamed KM, Van Dyke M, Rashed G, et al (2020)
Preliminary rib support requirements for solid coal
ribs using a coal pillar rib rating. In: Proceeding of the
39th International Conference on Ground Control in
Mining. Canonsburg, PA, USA, pp 85–96.
[2] Rashed G, Barton T, Sears M, et al (2018) Estimation
of the intact strength of coal using indirect methods.
In: Proceedings of the 37th International Conference
on Ground Control in Mining. Morgantown, WV,
USA, pp 294–301.
[3] Rusnak JA (2017) Coal strength variation by lithot-
ype for high-volatile A bituminous coal in the cen-
tral Appalachian Basin. In: Proceedings of the 36th
International Conference on Ground Control in
Mining. Morgantown, WV USA, pp 198–207.
[4] Singh A, Armstrong RT, Regenauer-Lieb K,
Mostaghimi P (2019) Rock Characterization Using
Gray-Level Co-Occurrence Matrix: An Objective
Perspective of Digital Rock Statistics. Water Resour Res
55:1912–1927. doi.org/10.1029/2018WR023342.
[5] Slaker BA, Mohamed KM (2016) A practical appli-
cation of photogrammetry to performing rib char-
acterization measurements in an underground coal
mine using a DSLR camera. In: Proceeding of the
35th International Conference on Ground Control
in Mining. pp 179–186.
[6] Stopes MC (1919) On the four visible ingredients in
banded bituminous coal. In: Proceedings of the Royal
Society of London. Series B, Containing Papers of a
Biological Character. pp 470–487.
[7] Sun J, Su B (2013) Coal–rock interface detection
on the basis of image texture features. Int J Min Sci
Technol 23:681–687.
[8] Xue Y (2022) Coal and Rock Classification with
Rib Images and Machine Learning Techniques.
Mining, Metall Explor 39:453–465. doi.org/10.1007
/s42461-021-00526-4.
[9] Yu K, Barry JC, Esterle JS (1997) Analysis of coal
banding texture and implications for megascopic
image analysis. Int J Coal Geol 33:1–18. doi.org
/10.1016/S0166-5162(96)00022-5