Institut für Lebensmittelwissenschaft und Biotechnologie

Permanent URI for this collectionhttps://hohpublica.uni-hohenheim.de/handle/123456789/6

Browse

Browsing Institut für Lebensmittelwissenschaft und Biotechnologie by Journal "Applied food research"

Type the first few letters and click on the Browse button
Now showing 1 - 2 of 2
  • Thumbnail Image
    Publication
    Influence of low oxygen concentrations on color stability of modified atmosphere packaged beef
    (2026) Krell, Johannes; Aeckerle, Luis; Poveda-Arteaga, Alejandro; Weiss, Jochen; Terjung, Nino; Gibis, Monika
    The influence of low oxygen concentrations on the development of color and the myoglobin redox states over storage time was analyzed, to determine whether there are conditions that increase discoloration. Beef slices were packaged in atmospheres containing nitrogen gas and 0 %, 0.5 %, 1 %, 1.5 %, 3 %, and 5 % of oxygen. The samples were stored at 2 °C for 14 days. During storage, color, reflectance and oxygen concentration were measured optically through the packaging. The color difference ΔE2000 and the relative oxymyoglobin (OMb), deoxymyoglobin (DMb), and metmyoglobin (MMb) levels were calculated. After 14 days, the oxygen concentrations changed to 0.09 % (0 %), 0.36 % (0.5 %), 0.92 % (1 %), 1.28 % (1.5 %) 2.55 % (3 %), and 4.29 % (5 %). Regarding MMb formation, the 0 % samples (ΔMMb0–14d 11.1 %) were significantly (p < 0.05) more stable compared to the other samples, which showed an increase of MMb formation with rising oxygen concentration after 14 days. The other samples reached a ΔMMb0–14d increase of 21.1 % (0.5 %), 26.7 % (1 %), 30.0 % (1.5 %), 31.1 % (3 %), and 34.4 % (5 %). The color stability showed significantly (p < 0.05) increasing ΔE values of 2.49 (0 %), 3.39 (0.5 %), 4.66 (1 %), 5.14 (1.5 %), 6.03 (3 %), and 7.34 (5 %) with rising oxygen contents. These findings suggest that to ensure the color stability of beef with minimal MMb formation, it is important to completely exclude oxygen from the packages, since the destabilizing effect of oxygen already started at 0.5 %. The non-invasive measurement of the oxygen concentration and the reflectance data over 14 days gave new insights into the discoloration process of beef stored in low-oxygen atmospheres.
  • Thumbnail Image
    Publication
    Poultry perfection : comparison of computer vision models to detect and classify poultry products in a production setting
    (2025) Einsiedel, Daniel; Vita, Marco; Jox, Dana; Dunnewind, Bertus; Meulendijks, Johan; Krupitzer, Christian
    This study explores the use of computer vision, specifically object detection, for quality control in ready-to-eat meat products. We focused on a single process step, labeling products as “good” or “imperfect”. An “imperfect product” constitutes a product that deviates from the norm regarding shape, size, or color (having a hole, missing edges, dark particles, etc.). Imperfect does not mean the product is inedible or a risk to food safety, but it affects the overall product quality. Various object detectors, such as YOLO, including YOLO12, were compared using the mAP50-95 metric. Most models achieved mAP scores over 0.9, with YOLO12 reaching a peak score of 0.9359. The precision and recall curves indicated that the model learned the “imperfect product” class better, most likely due to its higher representation. This underscores the importance of a balanced dataset, which is challenging to achieve in real-world settings. The confusion matrix revealed false positives, suggesting that increasing dataset volume or hyperparameter tuning could help. However, increasing the dataset volume is usually the more difficult path since data acquisition and especially labeling are by far the most time-consuming steps of the whole process. Overall, current models can be applied to quality control tasks with some margin of error. Our experiments show that high-quality, consistently labeled datasets are potentially more important than the choice of the model for achieving good results. The applied hyperparameter tuning on the YOLO12 model did not outperform the default model in this case. Future work could involve training models on a multi-class dataset with hyperparameter optimization. A multi-class dataset could contain more specific classes than just “good” and “imperfect,” making trained models capable of actually predicting specific quality deviations.