In breeding ducks, obtaining the pose information is vital for perceiving their physiological health, ensuring welfare in breeding, and monitoring environmental comfort. This paper proposes a pose estimation method by combining HRNet and CBAM to achieve automatic and accurate detection of duck's multi-poses. Through comparison, HRNet-32 is identified as the optimal option for duck pose estimation. Based on this, multiple CBAM modules are densely embedded into the HRNet-32 network to obtain the pose estimation model based on HRNet-32-CBAM, realizing accurate detection and correlation of eight keypoints across six different behaviors. Furthermore, the model's generalization ability is tested under different illumination conditions, and the model's comprehensive detection abilities are evaluated on Cherry Valley ducklings of 12 and 24 days of age. Moreover, this model is compared with mainstream pose estimation methods to reveal its advantages and disadvantages, and its real-time performance is tested using images of 256 × 256, 512 × 512, and 728 × 728 pixel sizes. The experimental results indicate that for the duck pose estimation dataset, the proposed method achieves an average precision (AP) of 0.943, which has a strong generalization ability and can achieve real-time estimation of the duck's multi-poses under different ages, breeds, and farming modes. This study can provide a technical reference and a basis for the intelligent farming of poultry animals.
Deep eutectic solvents (DESs) composed by amino acids (L-arginine, L-proline, L-alanine) as the hydrogen bond acceptors (HBAs) and carboxylic acids (formic acid, acetic acid, lactic acid, levulinic acid) as hydrogen bond donors (HBDs) were prepared and used for the dissolution of dealkaline lignin (DAL). The mechanism of lignin dissolution in DESs was explored at molecular level by combining the analysis of Kamlet-Taft (K-T) solvatochromic parameters, FTIR spectrum and density functional theory (DFT) calculations of DESs. Firstly, it was found that the formation of new hydrogen bonds between lignin and DESs mainly drove the dissolution of lignin, which were accompanied by the erosion of hydrogen bond networks in both lignin and DESs. The nature of hydrogen bond network within DESs was fundamentally determined by the type and number of functional groups in both HBA and HBD, which affected its ability to form hydrogen bond with lignin. One hydroxyl group and carboxyl group in HBDs provided active protons, which facilitated proton-catalyzed cleavage of β-O-4, thus enhancing the dissolution of DESs. The superfluous functional group resulted in more extensive and stronger hydrogen bond network in the DESs, thus decreasing the lignin dissolving ability. Moreover, it was found that lignin solubility had a closed positive correlation with the subtraction value of α and β (net hydrogen donating ability) of DESs. Among all the investigated DESs, L-alanine/formic acid (1:3) with the strong hydrogen-bond donating ability (acidity), weak hydrogen-bond accepting ability (basicity) and small steric-hindrance effect showed the best lignin dissolving ability (23.99 wt%, 60 °C). On top of that, the value of α and β of L-proline/carboxylic acids DESs showed some positive correlation with the global electrostatic potential (ESP) maxima and minima of the corresponding DESs respectively, indicating the analysis of ESP quantitative distributions of DESs could be an effective tool for DESs screening and design for lignin dissolution as well as other applications.