Imaging techniques and associated apparatus

1. A method of viewing compositional inhomogeneities in a solid chemical sample having a surface and a thickness, the method comprising:

irradiating the sample with a first pulse of radiation having a plurality of frequencies in the range from 25 GHz to 100 THz;

directing reflected and/or transmitted radiation from the sample onto a detector for combining with a second a second pulse of radiation, wherein at least one of the first and second pulses is delayed by a scanning delay line;

detecting radiation with the detector, over time to obtain a time domain waveform, wherein variation of the scanning delay allows for measurement of a delay time, wherein the delay time, which forms an x-axis of the time domain waveform, relates to the depth in the sample at which the pulsed radiation beam was reflected, wherein the time domain signal is determined corresponding to a range of depths in a region of interest within the sample;

selecting a depth of the sample to analyze;

converting the time domain waveform, localized to the selected depth, from the time domain to the frequency domain using a time domain to frequency domain transform to obtain frequency data localized to the selected depth; and

determining the composition of all or part of the sample using the frequency data for the selected depth.

2. The method according to claim 1 wherein the sample is a pharmaceutical sample.

3. The method according to claim 1 wherein the composition of all or part of the sample is determined from an image generated by plotting the frequency data for the selected depth, wherein the image is a three dimensional representation of compositional distribution.

4. The method according to claim 1 further comprising:

subdividing the sample into a two-dimensional array of pixels, detecting radiation from each pixel;

obtaining a time domain waveform for each pixels;

obtaining frequency data as a function of time for each pixel from the respective time domain waveforms;

deriving an image as a function of depth at each pixel from the respective frequency data; and

combining the images for each pixel into a three dimensional compositional distribution image for the sample.

5. The method according to claim 1 further comprising:

subdividing the sample into a two-dimensional array of pixels,

detecting radiation from each pixel;

obtaining a time domain waveform for each pixel;

obtaining frequency data as a function of time for each pixel from the respective time domain waveforms; and

deriving a cross-sectional compositional image from the respective frequency data.

6. The method according to claim 1 as used in a pharmaceutical manufacturing process.

7. The method of claim 1 , wherein determining the composition of all or part of the sample, comprises generating a representation of the composition of the sample, by plotting the frequency data for the selected depth.

8. The method of claim 1 , wherein the pulses are reflected by at least a first mirror and a second mirror before passing into the beam splitter.

9. The method of claim 8 , wherein the radiation is reflected by at least a third mirror and a fourth mirror before being directed onto the detector.

10. A method of viewing a granularity of a solid chemical sample having a surface and thickness, the method comprising:

irradiating the sample with a first pulse of radiation having a plurality of frequencies in the range from 25 GHz to 100 THz

directing the reflected and/or transmitted radiation from the sample onto a detector for combining with a second pulse of radiation, wherein at least one of the first and second pulses is delayed by a scanning delay line;

detecting radiation, with the detector, over time to obtain a time domain waveform, wherein variation of the scanning delay allows for measurement of a delay time, wherein the delay time, which forms an x-axis of the time domain waveform, relates to the depth in the sample at which the radiation was reflected;

selecting a depth of the sample to analyze;

converting the time domain waveform, localized to the selected depth, from the time domain to the frequency domain using a time domain to frequency domain transform to obtain frequency data localized to the selected depth; and

determining the granularity of all or part of the sample using the frequency data for the selected depth.

11. The method of claim 10 , wherein determining the granularity of all or part of the sample, comprises generating a representation of the composition of the sample, by plotting the frequency data for the selected depth.

12. The method of claim 10 , wherein the pulses are reflected by at least a first mirror and a second mirror before passing into the beam splitter.

13. The method of claim 12 , wherein the radiation is reflected by at least a third mirror and a fourth mirror before being directed onto the detector.

14. A method of viewing compositional inhomogeneities in a solid chemical sample having a surface and a thickness, the method comprising:

irradiating the sample with a first pulse of radiation having a plurality of frequencies in the range from 25 GHz to 100 THz

directing the reflected and/or transmitted radiation from the sample onto a detector for combining with a second pulse of radiation, wherein at least one of the first and second pulses is delayed by a scanning delay line;

detecting radiation, with the detector, over time to obtain a time domain waveform, wherein variation of the scanning delay allows for measurement of a delay time, wherein the delay time, which forms an x-axis of the time domain waveform, relates to the depth in the sample at which the pulsed radiation beam was reflected, wherein the time domain signal is determined corresponding to a range of depths in a region of interest within the sample;

selecting a depth of the sample to analyze;

converting the time domain signal, localized to the selected depth, from the time domain to the frequency domain using a Gabor transform to obtain frequency data localized to the selected depth; and

determining the granularity of all or part of the sample using the frequency data for the selected depth.

15. The method of claim 14 wherein the Gabor transform is implemented using a windowed Fourier transform, a correlation of a specific kernel function or a filter-bank.

16. The method of claim 14 further comprising applying the Gabor function to the time domain waveform and selecting frequency, window type and/or window width of the Gabor function to optimise spectral or temporal features.

17. The method of claim 14 , wherein determining the granularity of all or part of the sample, comprises generating a representation of the composition of the sample, by plotting the frequency data for the selected depth.

18. A method of viewing compositional inhomogeneities in a solid chemical sample having a surface and a thickness, the method comprising:

irradiating the sample with a first pulse of radiation having a plurality of frequencies in the range from 25 GHz to 100 THz

directing reflected and/or transmitted radiation from the sample onto a detector for combining with a second pulse of radiation, wherein at least one of the first and second pulses is delayed by a scanning delay line;

detecting radiation, with the detector, over time to obtain a time domain waveform, wherein variation of the scanning delay allows for measurement of a delay time, wherein the delay time, which forms an x-axis of the time domain waveform, relates to the depth in the sample at which the pulsed radiation beam was reflected, wherein the time domain signal is determined corresponding to a range of depth in a region of interest within the sample;

selecting a depth of the sample to analyze;

applying a window to the time domain waveform at the selected depth and converting the windowed time domain waveform to the frequency domain using a time domain to frequency domain transform to obtain frequency data localized to the selected depth; and

determining the composition of all or part of the sample using the frequency data for the selected depth.

19. The method of claim 18 , wherein determining the composition of all or part of the sample, comprises generating a representation of the composition of the sample, by plotting the frequency data for the selected depth.