Detect blemishes and bruising under the skin, define ripeness and chemical quality independent of the fruit color and size, find and remove foreign materials like plastic, wood, paper, metal, or insects. Specim FX cameras can reveal much more than traditional color and filter cameras or point spectrometers.

Achieve better quality, ripe products with optimized shelf life, and reduce losses and waste.

Integrate Specim FX Cameras to your quality control and sorting processes to enhance their capability in product inspection and process analytics. Non-invasive, real-time inspection replaces visual inspections and time-consuming lab tests. Covers 100% of the product stream. Improves chemical grading and foreign material detection. Detects a range of characteristics simultaneously with a single camera – now and in the future.

  • Non-invasive, real-time inspection replaces visual inspections and time-consuming lab tests.

  • Covers 100% of the product stream.

  • Improves chemical grading and foreign material detection.

  • Detects a range of characteristics simultaneously with a single camera – now and in the future.

List of quality parameters that can be detected with hyperspectral imaging.

Find the right camera for your needs

Different materials reflect light differently and therefore are visible on different wavelengths. Different cameras operate on different wavelengths, so finding the right solution depends on what do you wish to find. Specim has a wide variety of cameras to suit different needs; for example, FX10 can be used to determine if the apples are ripe or if they have bruises. If you want to know the sugar level on berries or potatoes, however, you need Specim FX17.

Find a compatible commercial software tool or use Specim SDK and ASCII control protocol to integrate the FX cameras to your own system software solutions. Do not hesitate to contact us if you need assistance, we are happy to help you!

Case example: Avocados

Annual production of avocados reached 5.5 million tons in 2016 – it has nearly doubled in 15 years thanks to its dietary value and health benefits. Avocado fruit does not ripen until detached from the tree and harvesting the fruit at optimum maturity is important to ensure the optimal eating quality. On one hand, the price of the fruit is higher at the beginning of the picking season, therefore the harvesters are tempted to pick immature fruits that do not ripen properly. As a result, the eating quality and the dietary value of the fruit suffers. On the other hand, an over-mature fruit picked at a late phase of harvesting period has a shorter shelf life and added risk of disorders and diseases.

Predicting the avocado eating quality and commercial value is difficult due to is non-apparent visual change. However, Specim sensors can detect and predict changes that are not visible to the eye. The changes in fruit quality have a specific spectral fingerprint that can be detected with Specim sensors. Prediction models can be built and applied to detect the fruits and their parts that are at higher risk to get blemished.

Spectral imaging shows blemishes under avocado’s skin. Picture courtesy of LuxFlux.

Why Specim

We have developed hyperspectral cameras specifically for industrial and production line use. They are small and lightweight, fast, and reliable, and the standard interfaces make it possible to integrate them to existing solutions.

Our products are called push-broom, or line-devices – they collect the full spectral and image information with single image. The final datacube is formed by scanning the target line-by-line. This is why you only need to illuminate that one single line where the data is collected.

Current main competing technologies are tunable filter and variable filter instruments. With tunable filter, one wavelength band is measured from two-dimensional area at a time. With linear variable filter, all wavelength bands are measured simultaneously but each band from different position on the target area.

Compared to push-broom, both tunable filter and variable filter technologies require homogenous and stable illumination for much wider two-dimensional area. They also struggle in getting co-registered spectrum, which leads to difficulties in data processing, unreliable spectral signatures and delayed processing results.

It is reasonable to conclude that only line imaging push-broom HSI is suitable to all industrial on-line applications where moving and changing targets are measured, and spectral co-registration is required.