What's the difference between multispectral and hyperspectral?

As an important means of natural science analysis, spectral analysis is often used to detect the physical structure, chemical composition and other indicators of objects. Image spectral measurement is the combination of spectral technology and imaging technology, which combines spectral resolution and graphic resolution to create the spatial dimension of surface spectral analysis, that is, the current multi-spectral imaging and hyperspectral imaging technologies.

Multispectral and hyperspectral are spectral imaging types with similar technology and application direction, so many people can't tell which is hyperspectral. Which one is multispectral?

So what's the difference between multispectral and hyperspectral?

First, the number of bands is different.

Multispectral images usually refer to 3 to 10 band; Hyperspectral images may have hundreds or thousands of bands.

Second, the spectral resolution difference.

The spectral resolution of multi-spectrum is poor, and the number that can be captured is relatively small because of the wide band; Hyperspectrum is composed of a narrower band (10-20 nm), with higher spectral resolution, which can detect the spectral effect of objects and provide more intangible data.

Third, the difference of information.

The low information content of multi-spectral images is the main reason that hinders their sustainable development. Hyperspectral images are rich in information and have great application and development space.

Fourth, the complexity is different.

Due to the limitation of the number of bands, multispectral is less complicated and easier to understand and apply, while hyperspectral needs more work to deal with.

Fifth, the image characteristics are different.

Hyperspectrum can present hundreds of points in each band, and more details can be observed, but multispectral can not be realized.

Sixth, camera differences.

Hyperspectral cameras can measure different wavelengths, covering infrared and ultraviolet regions, while multispectral cameras can only separate specific wavelengths.

Seventh, the cost is different.

Multi-spectrum only needs to collect several spectral bands, and the technology is not complicated, and the purchase and maintenance costs are low. However, the technical characteristics of hyperspectral need better and more technologies to support, and the cost is high.

Eighth, pixel synthesis difference.

Multispectral spectrum is a discrete sample spectrum, and each pixel may have 4 to 20 data points, while each pixel of hyperspectral spectrum is a continuous or complete spectrum.

Ninth, the treatment methods are different.

Multispectral processing defines images, while hyperspectral processing is spectra and images, which can present more dimensions.