This website uses cookies primarily for visitor analytics. Certain pages will ask you to fill in contact details to receive additional information. On these pages you have the option of having the site log your details for future visits. Indicating you want the site to remember your details will place a cookie on your device. To view our full cookie policy, please click here. You can also view it at any time by going to our Contact Us page.

Hyperspectral imaging is coming of age!

28 November 2016

Rob Webb explains why you may soon be hearing more about hyperspectral imaging, or chemical colour imaging.

Due to the complexity of analysis required hyperspectral imaging has not been used extensively in industrial applications. The technique makes use of the fact that organic materials selectively absorb light at different wavelengths in the infrared region of the spectrum depending on their composition. This gives distinctive ‘fingerprints’ which can be used to uniquely identify them. 

In hyperspectral imaging, a series of images are built up by sequentially allowing narrow wavelength bands of IR light from the sample to fall on the sensor. These images are combined to form a three-dimensional hyperspectral data cube. This data cube contains all the information needed to extract the chemical composition at each pixel. Historically, this huge amount of information has required a chemist and/or mathematician to crunch the data and understand the results, making it so costly that it has only been used in a handful of applications such as waste sorting for recycling.

A new approach 
A spectrograph is combined with a camera that is sensitive to IR wavelengths. The spectrograph allows the reflected light from the sample to be sorted into its constituent wavelengths. What has revolutionised the technique has been the emergence of affordable, flexible, high-speed data processing software such as Perception Studio from Perception Park. This new approach is called ‘chemical colour imaging’ and makes chemical material properties accessible to the machine vision engineer. 

The software engine extracts data from the complex data cube which is processed in real time to produce an image where the output of each pixel is colour coded according to the chemical composition of the material it is looking at. This is invaluable because the system reveals information that other machine vision technologies cannot. 

Rather than processing data from the entire spectrum, it may be possible to ‘tune into’ just a few key wavelengths that would distinguish between specific materials or identify known contaminants that could arise from a production processing stage. By reducing the data processing required, the image acquisition speed can be increased to reduce the overall inspection time. The technique has potential for use in a number of industries including food and pharmaceutical.

The data processing software turns the camera system into an easy-to-understand and intuitive configurable ‘chemical colour camera’. It offers a user interface to define the acquisition parameters and acquire the initial image data, explore the hyperspectral data with images and graphical feedback, design or define the best models for extracting the key information and then output colour coded images over a GigE Vision compliant stream in a standardised machine vision form. Online processing using standard image processing methods, such as grey level analysis and blob detection, can be used for colour sorting as in standard machine vision applications.

Chemical colour imaging in action
The potential of hyperspectral imaging can be illustrated in examples from the food and pharmaceutical industries. In a pistashio sorting application, for example (a) shows a traditional colour image of a mixture of pistachio nuts, shells and skin. Although some colour differences can be seen, they cannot be used to identify the different components. However, image (b) shows the hyperspectral image showing the nuts as red, the shells blue and the skin green. Not only are the component parts clearly distinguished for sorting it is also possible to differentiate those nuts that still have some skin attached.

Another benefit of hyperspectral imaging is that a lot of packaging material is transparent to the IR light meaning that the technique can be used to examine product inside its packaging. Cross-contamination in a pharmaceutical packing line could have potentially life-threatening consequences for the patient. The implications for the manufacturer could also be massive, both in terms of reputation and in terms of costly product recalls, and possible production line closure while the problem is investigated. 

New technologies are now making chemical colour imaging a practical reality, both in terms of costs and accessibility.

Rob Webb is a technical specialist at Stemmer Imaging. 


Contact Details and Archive...

Related Articles...

Most Viewed Articles...

Print this page | E-mail this page