Using Thermal Infrared Multispectral for the Imaging of Minerals

Multispectral imaging involves acquisition of multiple images of the same scene utilizing different spectral filters, representing a great compromise between hyperspectral imaging and broadband imaging. Spectral data can be acquired from the response of individual spectral filters, subtractions, ratios, and/or combinations of multiple filters. Although thermal infrared multispectral imaging is advantageous in many ways, the low acquisition rate or spectral band rate of commercial systems make them less attractive when compared to traditional imaging systems. The lack of rotation of the filter wheel is another issue in most of the systems because it is necessary to perform the selection and/or spectral filter change in each acquisition sequence. Such operation mode is laborious. Additionally, the limited availability of spectral filters (typically 4) leads to low spectral resolution, thus low selectivity. This article discusses the application of Telops MS-IR infrared camera to perform time-resolved multispectral imaging of minerals.

The Telops MS-IR Infrared Camera Series

The Telops MS-IR infrared camera series (Figure 1) equipped with a fast-rotating filter wheel conducts dynamic multispectral imaging on eight channels at a high frame rate. By synchronizing the FPA clocking and the filter wheel rotation, time-resolved multispectral imaging can be performed by recording a single frame at each filter position. This is followed by the calibration of sequences frame by frame utilizing in-band photon radiance (IBR) format in accordance with their respective spectral filter dataset.

Figure 1. Telops MS-IR infrared camera

Different models of MS-IR infrared cameras are available from Telops, covering the whole mid-infrared spectral range. The MS-IR VLW (very long wave 7.7-11.8µm) and MS-IR MW (midwave, 3.0-4.9µm) use 320x256 pixels MCT (Mercury-Cadmium-Telluride) and 640x512 pixels InSb focal plane array (FPA) detectors, respectively. They were held in oven set for few minutes at 150°C and then the minerals of interest were fed into the table, followed by recording of multispectral sequences.

The MS-IR-HD is equipped with a high-definition 1280x1024 pixels FPA detector, whereas the MS-IR-FAST uses a fast 320x256 pixels FPA detector to acquire images at the fastest frame rate available. The MS-IR infrared cameras provide spectral information about the targets of interest by splitting the scene radiance into eight different spectral bands. The fast rotating filter wheel mechanism increases the frame rate of the camera and is operable in both fixed and rotating modes. It provides a maximum effective frame rate of 800 Hz (100Hz per channel) by achieving a rotation speed of up to 6000rpm. All the MS-IR series cameras have the real-time radiometric and non-uniformity correction features based on Telops patented blackbody free correction technique.

Instrumentation and Experimental Procedure

This experiment used eight different filters to acquire spectral information. A neutral density filter occupied channel #1 and the associated frames represent broadband images. Depending on the spectral filter, acquisitions were performed utilizing the full FPA frame (320x256 pixels) at an integration time of 190- 400µs. The rotation speed set for the filter wheel was 1800rpm, resulting in an effective frame rate of 30 Hz/filter. All experiments used a circular 50mm Janos Varia lens. The spatial resolution achieved was 4mm2/pixel by placing the camera at a distance of 2m from the targets. Amethyst minerals, an iron pyrite (FeS2), and a hematite (Fe2O3) drill core contaminated with quartz (SiO2) veins were heated in oven for few minutes at 150°C. The minerals were then mounted on a table and multispectral sequences were subsequently recorded.

Figure 2. Visible image of a hematite drill core contaminated with quartz veins (left), iron pyrite (center) and amethyst (right).

The radiance transmission by spectral filters is typically over relatively wide spectral ranges. Radiometric calibration of multispectral cameras involves characterization of the detector and the responses of its optical components against known radiance values. Hence, the IBR procedure mostly involves integration of the Planck curve equation over a finite spectral range. It is possible to estimate the IBR of a given target for each filter as per its spectral emissivity for known concentration and thermal contrast conditions, thereby enabling the improvement of the thermal contrast for a selected target through correlation of its estimated IBR profile with measured IBR profiles of individual pixels in a scene.

Experimental Results

Selective Absorption/Emission of IR Radiation

Figure 3 shows the emissivity spectrum of quartz and each spectral filter’s transmittance curves convolved with the spectral response of the FPA detectors. In the 9.6 - 12 µm range, quartz is anticipated to act like a blackbody source and its self-emission will be function of its temperature as described by the Planck equation. The most potent spectral bands for quartz characterization can be assumed to be relative to filters #2, #4 and #5. Figure 4 illustrates the individual response of each acquisition channel on a normalized scale. Normalizing each IBR with the corresponding IBR of a blackbody source at an arbitrary temperature of 90°C was done for comparison of the response of each individual filter with each other. This normalization procedure facilitates evaluating the potential of each spectral filter in the characterization of a specific target on a mutual basis.

Figure 3. LWIR spectral emissivity of quartz (dark blue curve). The transmittance curves of each spectral filter used for the experiment are shown for evaluation purposes.

Figure 4. Normalized responses (blackbody source of 90 °C) of each acquisition channel for multispectral imaging experiment on minerals.

The infrared images suggest the sparse dispersion of the quartz across the entire hematite drill core. A same kind of spectral filter response trend is seen for the amethyst mineral other than filter #2, wherein the signal is greater than expected. The substantial difference in temperature between the amethyst mineral and the hematite drill core is illustrated in Figure 5.

Figure 5. Radiometric temperature image (top) of the experimental setup on minerals and the corresponding correlation images (bottom) obtained from the IBR profile of quartz.

In-band Photon Radiance

The detector response in a single channel is a function of different parameters and is also influenced by interfering agents. The contrast in a scene corresponding to a target of interest can be improved by performing image correlation of its IBR profile. To demonstrate this procedure, the IBR profile of quartz mineral and a blackbody source were computed at 90°C in accordance with their spectral emissivity features, and the results are presented in Figure 6.

Figure 6. In-band radiance (IBR) profiles estimated for a blackbody source (top) and a warm quartz mineral (middle) both at 90 °C as well as the measured IBR profile of selected pixels associated with quartz mineral (bottom).

A correlation algorithm is used to improve the dissimilarities between the IBR profiles. The correlation image obtained from the IBR profile of quartz is also in Figure 5. The contrasts in the image represent the spatial distribution of quartz throughout the minerals. As expected, a strong correlation is observed between the amethyst crystals at the surface of the mineral and the quartz IBR profile. The lower correlation for the iron pyrite mineral demonstrates the selectivity of the multispectral imaging over broadband imaging. The spectral data acquired by multispectral imaging enables contrast improvements based on both self-emission and spectral emissivity.

Conclusion

Time-resolved multispectral imaging enables characterizing solid targets like minerals more efficiently and in less time. IBR profiles derived from eight acquisition channels yield meaningful data for image contrast enhancement. The additional data obtained from dynamic multispectral imaging over traditional thermal cameras provides new opportunities for infrared signature measurements.

This information has been sourced, reviewed and adapted from materials provided by Telops Inc.

For more information on this source, please visit Telops Inc.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Telops Inc.. (2019, September 13). Using Thermal Infrared Multispectral for the Imaging of Minerals. AZoOptics. Retrieved on July 12, 2020 from https://www.azooptics.com/Article.aspx?ArticleID=937.

  • MLA

    Telops Inc.. "Using Thermal Infrared Multispectral for the Imaging of Minerals". AZoOptics. 12 July 2020. <https://www.azooptics.com/Article.aspx?ArticleID=937>.

  • Chicago

    Telops Inc.. "Using Thermal Infrared Multispectral for the Imaging of Minerals". AZoOptics. https://www.azooptics.com/Article.aspx?ArticleID=937. (accessed July 12, 2020).

  • Harvard

    Telops Inc.. 2019. Using Thermal Infrared Multispectral for the Imaging of Minerals. AZoOptics, viewed 12 July 2020, https://www.azooptics.com/Article.aspx?ArticleID=937.

Ask A Question

Do you have a question you'd like to ask regarding this article?

Leave your feedback
Submit