Stocking density presents a key factor affecting livestock and poultry production on a large scale as well as animal welfare. However, the current manual counting method used in the hemp duck breeding industry is inefficient, costly in labor, less accurate, and prone to double counting and omission. In this regard, this paper uses deep learning algorithms to achieve real-time monitoring of the number of dense hemp duck flocks and to promote the development of the intelligent farming industry. We constructed a new large-scale hemp duck object detection image dataset, which contains 1500 hemp duck object detection full-body frame labeling and head-only frame labeling. In addition, this paper proposes an improved attention mechanism YOLOv7 algorithm, CBAM-YOLOv7, adding three CBAM modules to the backbone network of YOLOv7 to improve the network’s ability to extract features and introducing SE-YOLOv7 and ECA-YOLOv7 for comparison experiments. The experimental results show that CBAM-YOLOv7 had higher precision, and the recall, mAP@0.5, and mAP@0.5:0.95 were slightly improved. The evaluation index value of CBAM-YOLOv7 improved more than those of SE-YOLOv7 and ECA-YOLOv7. In addition, we also conducted a comparison test between the two labeling methods and found that the head-only labeling method led to the loss of a high volume of feature information, and the full-body frame labeling method demonstrated a better detection effect. The results of the algorithm performance evaluation show that the intelligent hemp duck counting method proposed in this paper is feasible and can promote the development of smart reliable automated duck counting.

At present, fish farming still uses manual identification methods. With the rapid development of deep learning, the application of computer vision in agriculture and farming to achieve agricultural intelligence has become a current research hotspot. We explored the use of facial recognition in fish. We collected and produced a fish identification dataset with 3412 images and a fish object detection dataset with 2320 images. A rotating box is proposed to detect fish, which avoids the problem where the traditional object detection produces a large number of redundant regions and affects the recognition accuracy. A self-SE module and a fish face recognition network (FFRNet) are proposed to implement the fish face identification task. The experiments proved that our model has an accuracy rate of over 90% and an FPS of 200.

A video-based method to quantify animal posture movement is a powerful way to analyze animal behavior. Both humans and fish can judge the physiological state through the skeleton framework. However, it is challenging for farmers to judge the breeding state in the complex underwater environment. Therefore, images can be transmitted by the underwater camera and monitored by a computer vision model. However, it lacks datasets in artificial intelligence and is unable to train deep neural networks. The main contributions of this paper include: (1) the world’s first fish posture database is established. 10 key points of each fish are manually marked. The fish flock images were taken in the experimental tank and 1000 single fish images were separated from the fish flock. (2) A two-stage attitude estimation model is used to detect fish key points. The evaluation of the algorithm performance indicates the precision of detection reaches 90.61%, F1-score reaches 90%, and Fps also reaches 23.26. We made a preliminary exploration on the pose estimation of fish and provided a feasible idea for fish pose estimation.

The sex ratio is an important factor affecting the economic benefits of duck groups in the process of hemp duck breeding. However, the current manual counting method is inefficient, and the results are not always accurate. On the one hand, ducks are in constant motion, and on the other hand, the manual counting method relies on manpower; thus, it is difficult to avoid repeated and missed counts. In response to these problems, there is an urgent need for an efficient and accurate way of calculating the sex ratio of ducks to promote the farming industry. Detecting the sex ratio of ducks requires accurate counting of male ducks and female ducks. We established the world’s first manually marked sex classification dataset for hemp ducks, including 1663 images of duck groups; 17,090 images of whole, individual duck bodies; and 15,797 images of individual duck heads, which were manually captured and had sex information markers. Additionally, we used multiple deep neural network models for the target detection and sex classification of ducks. The average accuracy reached 98.68%, and with the combination of Yolov5 and VovNet_27slim, we achieved 99.29% accuracy, 98.60% F1 score, and 269.68 fps. The evaluation of the algorithm’s performance indicates that the automation method proposed in this paper is feasible for the sex classification of ducks in the farm environment, and is thus a feasible tool for sex ratio estimation.

Marine mammals are an important part of marine ecosystems, and human intervention seriously threatens their living environments. Few studies exist on the marine mammal call recognition task, and the accuracy of current research needs to improve. In this paper, a novel MG-ResFormer two-channel fusion network architecture is proposed, which can extract local features and global timing information from sound signals almost perfectly. Second, in the input stage of the model, we propose an improved acoustic feature energy fingerprint, which is different from the traditional single feature approach. This feature also contains frequency, energy, time sequence and other speech information and has a strong identity. Additionally, to achieve more reliable accuracy in the multiclass call recognition task, we propose a multigranular joint layer to capture the family and genus relationships between classes. In the experimental section, the proposed method is compared with the existing feature extraction methods and recognition methods. In addition, this paper also compares with the latest research, and the proposed method is the most advanced algorithm thus far. Ultimately, our proposed method achieves an accuracy of 99.39% in the marine mammal call recognition task.

Due to memory and computing resources limitations, deploying convolutional neural networks on embedded and mobile devices is challenging. However, the redundant use of the 1 × 1 convolution in traditional light-weight networks, such as MobileNetV1, has increased the computing time. By utilizing the 1 × 1 convolution that plays a vital role in extracting local features more effectively, a new lightweight network, named PlaneNet, is introduced. PlaneNet can improve the accuracy and reduce the numbers of parameters and multiply-accumulate operations (Madds). Our model is evaluated on classification and semantic segmentation tasks. In the classification tasks, the CIFAR-10, Caltech-101, and ImageNet2012 datasets are used. In the semantic segmentation task, PlaneNet is tested on the VOC2012 datasets. The experimental results demonstrate that PlaneNet (74.48%) can obtain higher accuracy than MobileNetV3-Large (73.99%) and GhostNet (72.87%) and achieves state-of-the-art performance with fewer network parameters in both tasks. In addition, compared with the existing models, it has reached the practical application level on mobile devices. The code of PlaneNet on GitHub: https://github.com/LinB203/planenet.