6. Conclusion

As this article has shown, there are many different instrumental principles in spectro-imaging, and it is clear that there is no such thing as a "universal" spectro-imager. We can, however, offer a few guidelines for anyone wishing to design or acquire a hyperspectral imager.

The first step is to define your needs as precisely as possible. The main criteria are spatial and spectral resolution, angle of field, and spectral range, but it may also be important to specify other parameters: signal-to-noise ratio, acquisition frequency, rejection rate and stray light rate, power consumption, etc.

With this need defined, the first point to look at is the link budget. If it appears that the available flux is sufficient, there's probably no need to turn to multiplexed measurement (Fourier transform, Hadamard transform or spectro-tomograph spectro-imagers),...