Moisture content is a crucial parameter in many industries:
- For the food industry, moisture strongly correlates with the product quality and freshness. It also needs to be monitored appropriately when one is producing dry fruits.
- Monitoring the moisture level of fuel (e.g., wooden ships) is crucial to optimize the burning process and efficiency.
- In the paper industry, water content needs to be monitored at many production stages to ensure end-product quality.
- In precision agriculture, moisture in leaves is a key indicator of plant health.
Monitoring moisture content increases production quality and optimizes processes in many industries. New agile and accurate technologies are needed, and within this context, Specim hyperspectral cameras present many assets.
Water has strong absorption features within the NIR spectral range, and the use of Specim FX17 hyperspectral camera is therefore natural and evident for monitoring moisture content. NIR spectroscopy combined with chemometric algorithms reveals the moisture level quantitatively. Since spectral imaging combines spectroscopy with imaging, the Specim FX17 would also map the spatial distribution of moisture, which is crucial for some applications (e.g., precision farming and meat processing).
In this study, we investigated minced meat samples’ moisture levels. The moisture level correlates tightly with freshness, and its level needs to be precisely monitored, especially before packaging. For this study, Atria a Finnish food industry company provided ten minced meat samples. To know their moisture level accurately, Specim ordered measurements from a 3rd party laboratory, certified in moisture analysis (Seilab in Seinäjoki, Finland; method NMKL 14:2012; See Table 1 below).
Measured value by Seilab | Measured value by FX17 | |
---|---|---|
Sample 1 | 75.8 | 76.4 |
Sample 2 | 72.6 | 72.3 |
Sample 3 * | 69.2 | 69.5 |
Sample 4 | 64.7 | 64.3 |
Sample 5 (mostly fat) | 17.9 | 17.8 |
Sample 6 | 74.9 | 74.7 |
Sample 7 | 72.8 | 71.9 |
Sample 8 | 68.2 | 68.8 |
Sample 9 * | 60.3 | 62.2 |
Sample 10 | 58.9 | 59.7 |
Table 1: moisture level (g/100g) on each sample included in this study. Samples 3 and 9 were used for validation purposes.
We measured the samples with Specim FX17 hyperspectral camera (Fig.1). It collects NIR spectra for each pixel of the acquired image (900 – 1700 nm). Those can be converted into moisture content employing a regression model. We used eight of these samples to build and calibrate the model. We used the two remaining samples for validation (indicated with * in Table 1).
Figure 1: FX17 on the 40×20 scanner (left) and example of a sample on the scanner sample tray (right).
The regression model results are presented in Table 1 and Fig.2. It clearly shows that the FX17 is a suitable tool to precisely measure minced meat’s moisture level.

Figure 2: regression plot of the quantitative model for moisture level prediction. Red dots relate to calibration samples, whereas green ones relate to validation samples.
In addition to measuring the moisture level in samples, hyperspectral imaging is suitable for measuring its distribution (Fig.3).

Figure 3: Example of Moisture distribution on a meat sample (here Sample 3).
Conclusions:
Specim FX17 camera provides integrators with crucial and accurate information about moisture level quantification. Besides, this fast and non-destructive method is also suitable for measuring other properties beneficial for process optimization or quality control. Similarly, the flexibility of hyperspectral imaging allows a rapid adaptation to new regulations and challenges.