Inspections are an important part of HACCP programs in the food industry (HACCP - Hazard Analysis Critical Control Point). However, metal detectors, X-ray and conventional image processing - important HACCP methods for detecting physical contaminants - have their limits. Metal detectors focus on the detection of metallic contaminants, while X-ray inspections are only effective if there is a sufficient difference between the density of the contaminant and that of the product being inspected. Conventional image processing systems search for contaminants using parameters such as size, shape and color.

Hyperspectral image processing is a powerful complementary concept: it combines spectroscopic analysis with image processing so that the images can be color-coded according to the chemical composition of the imaged substances. This means that both organic and inorganic impurities can be detected with one and the same system. The ability to detect differences in the chemical composition of organic substances opens up new possibilities. Systems are now available that work in real time so that they can be used in high-speed production lines.

The possibilities of hyperspectral image processing can be illustrated using an example with confectionery: jelly beans contain some non-metallic foreign bodies. These have different shapes and colors, but a similar density to the confectionery itself. Hyperspectral image processing makes them perfectly recognizable. The jelly beans all have the same chemical composition and are shown in green, while the impurities have a different chemical composition and appear in red.

The principle of color imaging of chemical substances is based on the spectroscopic analysis of the wavelength of the light coming from the object in real time. This produces a "chemical fingerprint" of the imaged substance. Each chemical component detected can be given its own color coding on the image. A spectrograph is needed to break down the light reflected from the sample into the different wavelengths of its components. An IR-sensitive sensor then creates an image of the sample for each of these wavelengths. These images are assembled into a three-dimensional hyperspectral data cube.

The CVS Hyperinspect system integrates these core components into a functional system that can be adapted to the specific requirements of each application. Flexible high-speed software extracts the data from the data cube and uses it to generate color assignments in the images. The images can be further processed and analyzed using the image processing software that is also included. CVS Hyperinspect simplifies the overall process of the hyperspectral method and opens up additional applications, for example in the pharmaceutical and packaging industries. Many packaging materials typically allow infrared light to pass through unhindered. This allows the chemical composition of the contents to be tested even through the packaging. as