The present invention relates to remote sensing. More specifically, the present invention relates to systems and methods for a hyperspectral imager that is suitable for gathering data related to specific gases.
The recent emphasis and concern about climate change has highlighted the need for better and more accurate tracking and assessment of gases in the atmosphere as produced by human activity. To this end, while remote sensing of gases such as methane is well-known and has been practiced for years, the methods and equipment used for such endeavors are sometimes considered as an afterthought.
Classic tracking and measurement of gases escaping from man-made facilities have been accomplished using equipment designed to detect multiple items/artifacts. Such hyperspectral imagers can be used for mapping vegetation, for archaeological sensing, detecting ores in the ground, and, of course, detecting gases. Such wide-ranging uses, while advantageous for manufacturers of the equipment, means that the equipment is not typically optimized for gas detection and characterization. Accordingly, the level of error and accuracy for gas detection and characterization may be less than ideal.
Based on the above, there is a need for systems and methods that reduce the amount of error and increases the accuracy of equipment used for hyperspectral imaging. Preferably, such systems and methods can be optimized for specific uses such as gas characterization and detection.
The present invention provides systems and methods relating to a multi-slit hyperspectral imager. The imager is configured with multiple slits that are parallel to one another. Each slit produces its own hyperspectral cube and is limited to a specific wavelength range. The multiple slits produce multiple data sets, obtained in quick succession, for the same section of an area to be imaged. In optical spectrometry applications such as trace gas sensing and quantification, this allows for improved measurement precision . The imager may be used for any gas of interest by adjusting the wavelength range to one that contains absorption features of the targeted gas.
In a first aspect, this document discloses a hyperspectral imager comprising:
In a second aspect, the present invention provides a hyperspectral imager comprising:
In one embodiment, the spectrally dispersive element is a grating. In another embodiment, the spectrally dispersive element is a prism.
In a further embodiment, the slit apertures are evenly spaced apart from one another. The split apertures may also be parallel to each other. As well, the split apertures may be normal to a direction of travel of said platform.
In another embodiment, the specific wavelengths of light include absorption wavelengths of a specific gas. The specific gas may be methane.
In another aspect, this document discloses a method of hyperspectral imaging, said method comprising:
In another embodiment, this document discloses a method wherein each slit aperture corresponds to a limited number of specific wavelengths of light received through said slit aperture.
In another embodiment, this document discloses a method wherein said hyperspectral imager further comprises a plurality of pixels for receiving split light on a focal plane, said split light resulting from light received through at least one of said plurality of slit apertures.
In another embodiment, this document discloses a method wherein each slit aperture corresponds to a fixed number of n pixels of said plurality of pixels, such that n < m, where m is a maximum number of available pixels.
In another embodiment, this document discloses a method wherein said multiple slit apertures are parallel to one another in a slit plane.
In another embodiment, this document discloses a method wherein said multiple slit apertures are evenly spaced apart from one another.
In another embodiment, this document discloses a method wherein said multiple slit apertures are normal to a direction of travel of said imaging platform.
In another embodiment, this document discloses a method wherein said specific wavelengths of light include absorption wavelengths of a specific gas.
In another embodiment, this document discloses a method wherein said specific gas is methane.
In another embodiment, this document discloses a method wherein said hyperspectral imager further comprises at least one spectrally dispersive element, said at least one spectrally dispersive element comprising at least one of a grating and a prism.
The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.
The embodiments of the present invention will now be described by reference to the following figures, in which identical reference numerals in different figures indicate identical elements and in which:
A conventional hyperspectral imaging system 10 is illustrated in
For a number of gas remote sensing applications, the desired performance metric is the detection threshold or the minimum detectable emission from a point source. For a fixed wind speed, this metric is driven primarily by column precision and spatial resolution (i.e., ground sampling distance). It should be clear that the term “column” refers to “column density” (a measure of gas abundance along the optical path). It should also be clear that “column precision” refers to the precision with which the column density is measured.
As will be shown below, the column precision can be improved by using a multi-slit configured hyperspectral imager to reduce the average noise. It should be clear that a multi-slit configured hyperspectral imager is a hyperspectral imager equipped with multiple slit apertures for receiving images of a scene to be imaged.
For the hyperspectral imager according to one aspect of the present invention, the spectral range is fixed to the wavelength range λ1 to λ2. For this implementation, this part of the spectrum contains absorption features associated with a specific gas of interest. As well, for this implementation, the spectrum gathered for each ground location is dispersed over n pixels of the focal plane array (the sensor 70 in
It should be clear that the precision of the estimate of the detected column density will depend on the per pixel signal levels, the per pixel noise levels, and the sensitivity of the signals to changes in the column density (χ).
Under certain conditions, the precision Δχ (mol/m2) of the column density estimate can be written as
where
Normalizing both the noise and the Jacobian above by the mean signal level gives the following expression:
This expression can now be evaluated under certain conditions/assumptions. Since the spectral range has been fixed to the range between λ1 and λ2, and for a given optical configuration, pixel array and integration time τ, it can be safely assumed that the total signal (or charge) summed over all pixels i is fixed (Q). Accordingly, for the case of n spectral pixels, 〈S〉 = Q/n and, in the shot-noise limit (σ = √S), this results in
Furthermore, without loss of generality, the fractional Jacobian norm can be expressed as :
where Δ
Based on the above, this results in
From this, if the spectrum is spread over fewer pixels n, the precision only starts to suffer when the sampling/broadening causes the Jacobian magnitude Δ
For a given set of circumstances (i.e., a given instrument, spectral range, and gas of interest), it is therefore possible that the precision may remain approximately invariant for n well below the total number of available spectral pixels N.
Because of the above, only a subset of the spectral pixels needs to be used. The rest of the spectral pixels can thus be used to repeat the measurement using multiple/additional slits. With additional repeated measurements of the same signal, the noise can be averaged down and, accordingly, column precision can be improved.
If multiple slits are used (e.g., k slits) and when the column error is random (and uncorrelated) for each successive measurement (as is the case with the shot-noise limit), the column precision will scale down as:
where the subscript refers to the number of slits.
From the above, it should be clear that, as k is increased, as long as the Jacobian rms Δ
The optimal value for k can be found by modeling and/or experimentation and such a value should provide the lowest value for Δχk.
Thus, in one aspect, the invention provides a hyperspectral imager configured with multiple parallel slits. The imager may be deployed on a platform that overflies an area to be imaged. Conventionally, each slit is configured to be at right angles to the direction of travel of the platform. As can be imagined, each slit images a section of a scene to be imaged. As the platform overflies the area, each section is repeatedly imaged as each successive slit passes over the section of the scene. This allows each section to be imaged repeatedly over a period of time. The data gathered for each image can then be correlated / aligned such that multiple data sets covering the same section can be overlaid, for example for averaging purposes. It should be clear that, ideally, the period of time over which the section is repeatedly imaged is short enough that an emission plume being imaged stays relatively stable. Preferably, local wind speed and direction do not change during this period of time during which the section is repeatedly imaged.
Referring to
For gas/plume related applications, as mentioned above, ideally the section of the scene to be imaged is repeatedly imaged over a period of time during which the plume stays relatively stable. As such, implementations in which the scene or a plume is repeatedly imaged over a period of approximately, for example, 15 seconds has been found to provide useful results.
It should be clear that the multiple slits allow the hyperspectral imager to operate as multiple, independent spectrometers, with each slit-based subsystem yielding a time-delayed dataset for the same section of the scene.
Referring to
It should also be clear that, as an example,
To illustrate the delay between the readings of the different slits in a multi-slit configured imager, suitable parameters for an implementation may be as follows for an aircraft deployed multi-slit imager with 10 evenly spaced slits:
The above parameters will result in an along-track speed of 88 pixels per second with a full spectral dimension of the focal plane array being traversed in 14.5 seconds. Assuming 10 equally/evenly spaced slits, this results in a delay of approximately 1.45 sec between the retrieval fields of successive slits. Thus, in 14.5 sec, 10 simultaneous readings of the swath are performed to result in better-precision spectrometry. It should be clear that the parameters given above simply provide an example implementation and other parameters are, of course, possible.
It should be clear that, while the above implementations discuss targeting methane plumes and methane emissions, other gases may also be targeted. Such targeting can be accomplished by merely adjusting the target frequency range for the hyperspectral datacubes.
Similarly, the multi-slit configured hyperspectral imager may be used for uses other than trace gas sensing and quantification. The multi-slit hyperspectral imager can be used in implementations or fields that can benefit from a better effective SNR. The concept of using multiple slit apertures in parallel may be useful in Raman spectroscopy as well as in LIBS spectroscopy (laser-induced breakdown spectroscopy).
It should be clear that, unless otherwise specified, any references herein to ‘image’ or to ‘images’ refer to a digital image or to digital images, comprising pixels or picture cells.
The construction “at least one of [x] and [y]”, as used herein, means and should be construed as meaning “[x], [y], or both [x] and [y]”.
A person understanding this invention may now conceive of alternative structures and embodiments or variations of the above all of which are intended to fall within the scope of the invention as defined in the claims that follow.
The present application claims priority to U.S. Provisional Application No. 63/336,655, filed on Apr. 29, 2022, the entirety of which is herein incorporated by reference.
Number | Date | Country | |
---|---|---|---|
63336655 | Apr 2022 | US |