Martin Georg Ljungqvist

Multispectral Image Analysis for Robust Prediction of Astaxanthin Coating

The aim of this study was to investigate the possibility of predicting the type and concentration level of astaxanthin coating of aquaculture feed pellets using multispectral image analysis. We used both natural and synthetic astaxanthin, and we used several different concentration levels of synthetic astaxanthin in combination with four different recipes of feed pellets. We used a VideometerLab with 20 spectral bands in the range of 385–1050 nm. We used linear discriminant analysis and sparse linear discriminant analysis for classification and variable selection. We used partial least squares regression (PLSR) for prediction of the concentration level. The results show that it is possible to predict the level of synthetic astaxanthin coating using PLSR on either the same recipe, or when calibrating on all recipes. The concentration prediction is adequate for screening for all recipes. Moreover, it shows that it is possible to predict the type of astaxanthin used in the coating using only ten spectral bands. Finally, the most selected spectral bands for astaxanthin prediction are in the visible range of the spectrum.

Image Analysis of Pellet Size for a Control System in Industrial Feed Production

When producing aquaculture fish feed pellets, the size of the output product is of immense importance. As the production method cannot produce pellets of constant and uniform size using constant machine settings, there is a demand for size control. Fish fed with feed pellets of improper size are prone to not grow as expected, which is undesirable to the aquaculture industry. In this paper an image analysis method is proposed for automatic size-monitoring of pellets. This is called granulometry and the method used here is based on the mathematical morphological opening operation. In the proposed method, no image object segmentation is needed. The results show that it is possible to extract a general size distribution from an image of piled disordered pellets representing both length and diameter of the pellets in combination as an area.

Near-infrared hyper-spectral image analysis of astaxanthin concentration in fish feed coating

The aim of this study was to investigate the possibility of predicting concentration levels of synthetic astaxanthin coating of aquaculture feed pellets by hyper-spectral image analysis in the near infra-red (NIR) range and optical filter design. The imaging devices used were a VideometerLab with 20 wavelengths in the range of 385-1050 nm, and a Specim camera with 256 wavelengths in the range of 970-2500 nm. Linear discriminant analysis (LDA), partial least squares regression (PLSR), and a modified stepwise random selection with ordinary least squares (OLS) for filter selection was used for prediction of the concentration level. The results show that it is possible to predict the level of synthetic astaxanthin coating using either hyper-spectral imaging or three bandpass filters (BPF).