Fermentation can increase the concentration of active ingredients and improve the effectiveness of Chinese herbal medicine. The purpose of this study was to investigate the effect of solid-state fermentation (SSF) on the polysaccharide content and color of Glycyrrhiza stems and leaves, as well as to explore the potential of computer vision-based analytical chemistry for the rapid, non-destructive, and accurate quality identification of fermented herbs. The effects of different inoculation rates on the polysaccharide content and color of fermented Glycyrrhiza stems and leaves were evaluated. Subsequently, dynamic changes in the viable counts of the probiotic strains, pH values, polysaccharide content, and color of Glycyrrhiza stems and leaves were explored during the entire fermentation process. The correlations of color variables that were extracted from the images with key quality indicators of the Glycyrrhiza stem and leaf samples were verified. The results showed that with an increase in inoculation amount, the polysaccharide content demonstrated a trend of first increasing and then decreasing, which was consistent with the color parameter behavior, and the optimal inoculation amount was 0.2%. During fermentation, R, G, B, S, V, L, a*, and B* were significantly correlated with the polysaccharide content (p < 0.01), while the correlation of H was weak. Principal component analysis (PCA) based on color variables can effectively distinguish between different stages of fermentation. This study provides a reference for the rapid and nondestructive analysis of fermented Glycyrrhiza stems and leaves, offering a new approach to process monitoring and quality control of fermented herbs.

This study introduces a non-contact, single-target method for real-time monitoring of dairy cow calving posture and behavior using the YOLOv8 model. In total, 600 videos were collected, from which 10,544 image samples were extracted through frame-by-frame processing. Complete video recordings of 86 cows (30 primiparous and 56 multiparous) were utilized to investigate changes in calving behavior. The YOLOv8 model achieved excellent performance with precision (P), recall (R), and mean average precision (mAP) of 96.72%, 96.53%, and 97.41%, respectively, and recognition P of 89.19% for lying postures and 82.61% for standing postures. Behavioral analysis revealed that lying postures were more frequent than standing, and primiparous cows had more frequent posture transitions (9.07 changes) than multiparous cows (5.29 changes), particularly during early parturition. Primiparous cows also showed significantly higher average times for parturition and lying as well ashigher frequency of behavioral changes compared to multiparous cows. Additionally, calf birth weight was positively correlated with maternal behaviors, especially in primiparous cows. Our proposed model effectively and accurately recognizes calving postures in dairy cows, enabling the early detection of abnormal calving events. This provides a scientific basis and technical support for intelligent farm management.

The ‘head back’ posture is a pronounced and significant behavioral trait during bovine parturition, commonly interpreted as a natural response to the pain associated with parturition. Leveraging computer vision technology for real-time monitoring of parturition behaviors can provide timely assistance during calving and enhance animal welfare. This study initially evaluated the head back posture in cows of different types, finding that primiparous cows and those delivering calves weighing over 43 kg exhibited prolonged durations of both labor and head back posture. A model was developed using the YOLOv8 algorithm with 25,617 images to recognize and classify changes in head posture during parturition, including positions like lying with or without head back. The model demonstrated robust predictive performance with a precision (P) of 69.76%, recall (R) of 75.35%, average precision (AP) of 70.12%, and F₁ score of 0.71. Furthermore, the model’s capability to recognize postures from different camera angles and under varying environmental conditions was assessed. Notably, images captured from an abdominal angle achieved AP exceeding 90%, with consistent stability under varying lighting conditions, including sunny and overcast weather, during both daytime and nighttime. Behavioral analysis showed that the parturition duration and total duration of head back posture in primiparous cows were significantly higher than those in multiparous cows (p < 0.05), and the changing trends of motor performance between primiparous and multiparous cows were consistent across different parturition stages. Additionally, the correlation between calf birth weight and maternal behavior was stronger in primiparous cows than in multiparous cows, further indicating obvious differences in physiological and behavioral responses of cows during primiparous and multiparous parturition. This study underscores the potential of computer vision applications in enhancing real-time intervention and promoting welfare during bovine parturition.

Heat stress is known to affect dairy cow behavior, but its effects on milking-order patterns and stabilization after herd entry in large commercial herds are not well described. We analyzed 70,491 parlor records from 484 Holstein cows milked three times daily in a free-stall herd to examine how thermal conditions shape milking-order dynamics. Daily intraday milking-order displacement was quantified based on changes in queue position between successive milkings. Cow-specific regressions of milking order on ambient temperature were used to identify clusters of thermal response, and time to reach a stable queue position was evaluated for newly introduced cows according to thermal conditions at herd entry. Results show a modest increase in intraday physical displacement from thermally comfortable to severely hot days, whereas relative displacement did not differ significantly across the four thermal categories. Cows showed consistent individual differences in how their milking position shifted with temperature that were only weakly associated with parity, days in milk, milk yield, or panting score. In an exploratory severe versus non-severe comparison, cows entering the herd under severe thermal conditions had a lower adjusted hazard of reaching the predefined stabilization criterion (HR = 0.62, 95% CI: 0.43–0.90). Collectively, these findings indicate that milking-order patterns extracted from routine parlor records are associated with behavioral responses to thermal conditions. Their management implications, including decisions related to herd entry and cooling, require further experimental evaluation.

Feeding behavior is closely related to nutrient intake, health, and production performance, yet its temporal dynamics are difficult to quantify in group-housed dairy goats. We developed a computer vision-based framework to characterize feeding dynamics under commercial farm conditions. Video data were collected from 550 Guanzhong dairy goats housed in five pens over three consecutive days. YOLO11, RF-DETR, and RT-DETR were compared for feeding-area goat detection, and temporal feeding features were analyzed to characterize high feeding activity (HFA) and low feeding activity (LFA). In a subsequent recording phase, head-down feeding behavior was assessed in 150 goats originating from the classified source pens. A total of 24 of these goats (12 HFA and 12 LFA) were selected for exploratory milk-production and serum comparisons. RF-DETR showed the best overall detection performance, with a precision (P) of 87.39%, recall (R) of 84.00%, F1-score of 85.66%, mean average precision (mAP)@0.5 of 90.38%, and mAP@0.5:0.95 of 53.30%. Goats exhibited a “rapid entry–peak aggregation–gradual departure” feeding pattern after feed delivery. Among the temporal features, rise time of feeding-area goat count (RTFGC), fall time of feeding-area goat count (FTFGC), and feeding-area goat count increase rate (FGCIR) were the most informative features for distinguishing HFA from LFA. HFA goats reached peak feeding-area occupancy earlier, entered the feeding area at a higher rate, and maintained feeding activity for longer than LFA goats. HFA goats also showed higher head-down feeding proportions, greater average daily dry matter intake (DMI), and higher milk yield, as well as distinct serum immune, antioxidant, and neuroendocrine-related profiles. These findings indicate that feeding dynamics derived from video observations can provide a quantitative digital behavioral phenotype for characterizing individual differences in dairy goat feeding activity and their associations with production and physiological status.