Application of automated image analysis to coal petrography

International Journal of Coal Geology
By: , and 



The coal petrologist seeks to determine the petrographic characteristics of organic and inorganic coal constituents and their lateral and vertical variations within a single coal bed or different coal beds of a particular coal field. Definitive descriptions of coal characteristics and coal facies provide the basis for interpretation of depositional environments, diagenetic changes, and burial history and determination of the degree of coalification or metamorphism. Numerous coal core or columnar samples must be studied in detail in order to adequately describe and define coal microlithotypes, lithotypes, and lithologic facies and their variations. The large amount of petrographic information required can be obtained rapidly and quantitatively by use of an automated image-analysis system (AIAS).

An AIAS can be used to generate quantitative megascopic and microscopic modal analyses for the lithologic units of an entire columnar section of a coal bed. In our scheme for megascopic analysis, distinctive bands 2 mm or more thick are first demarcated by visual inspection. These bands consist of either nearly pure microlithotypes or lithotypes such as vitrite/vitrain or fusite/fusain, or assemblages of microlithotypes. Megascopic analysis with the aid of the AIAS is next performed to determine volume percentages of vitrite, inertite, minerals, and microlithotype mixtures in bands 0.5 to 2 mm thick. The microlithotype mixtures are analyzed microscopically by use of the AIAS to determine their modal composition in terms of maceral and optically observable mineral components. Megascopic and microscopic data are combined to describe the coal unit quantitatively in terms of (V) for vitrite, (E) for liptite, (I) for inertite or fusite, (M) for mineral components other than iron sulfide, (S) for iron sulfide, and (VEIM) for the composition of the mixed phases (Xi) i = 1,2, etc. in terms of the maceral groups vitrinite V, exinite E, inertinite I, and optically observable mineral content M. The volume percentage of each component present is indicated by a subscript. For example, a lithologic unit was determined megascopically to have the composition (V)13(I)1(S)1(X1)83(X2)2. After microscopic analysis of the mixed phases, this composition was expressed as (V)13(I)1(S)1(V63E19I14M4)83(V67E11I13M9)2. Finally, these data were combined in a description of the bulk composition as V67E16I13M3S1. An AIAS can also analyze textural characteristics and can be used for quick and reliable determination of rank (reflectance).

Our AIAS is completely software based and incorporates a television (TV) camera that has optimum response characteristics in the range of reflectance less than 5%, making it particularly suitable for coal studies. Analysis of the digitized signal from the TV camera is controlled by a microprocessor having a resolution of 64 gray levels between full illumination and dark current. The processed image is reconverted for display on a TV monitor screen, on which selection of phases or features to be analyzed is readily controlled and edited by the operator through use of a lightpen.

We expect that automated image analysis, because it can rapidly provide a large amount of pertinent information, will play a major role in the advancement of coal petrography.

Publication type Article
Publication Subtype Journal Article
Title Application of automated image analysis to coal petrography
Series title International Journal of Coal Geology
DOI 10.1016/0166-5162(82)90002-7
Volume 2
Issue 2
Year Published 1982
Language English
Publisher Elsevier
Description 38 p.
First page 113
Last page 150
Google Analytic Metrics Metrics page
Additional publication details