This study aims to solve the problems of inaccurate weighing of bamboo sticks and inefficient manual counting. To overcome this problem, an improved MobileNetV3 model and a counting algorithm suitable for bamboo sticks—combined with a spatial-temporal attention mechanism—are proposed in this paper. Inspired by the idea of EfficientNet, scaling coefficients are used to scale the MobileNetV3 network structure as a whole in terms of width and height. The optimal model for bamboo-stick recognition is screened initially, and then the algorithm uses the convolutional block attention module (CBAM) attention mechanism to replace the squeeze-and-excitation (SE) attention mechanism in the MobileNetV3 network structure to allow the network to extract features in the two dimensions of channel and space. Since the number of bamboo sticks in a single image is extremely dense—generally around 1000–3000—it is difficult to effectively count them with existing algorithms. The proposed algorithm divides the image into multiple equally sized blocks and then uses the boundary processing algorithm to merge the cut bamboo stick images and count the number of sticks. Experimental results show that the proposed algorithm can effectively perform near real-time detection on a mobile terminal and its accuracy can reach approximately 97%, which is in line with actual production applications.

Traditional fish farming methods suffer from backward production, low efficiency, low yield, and environmental pollution. As a result of thorough research using deep learning technology, the industrial aquaculture model has experienced gradual maturation. A variety of complex factors makes it difficult to extract effective features, which results in less-than-good model performance. This paper proposes a fish detection method that combines a triple attention mechanism with a You Only Look Once (TAM-YOLO)model. In order to enhance the speed of model training, the process of data encapsulation incorporates positive sample matching. An exponential moving average (EMA) is incorporated into the training process to make the model more robust, and coordinate attention (CA) and a convolutional block attention module are integrated into the YOLOv5s backbone to enhance the feature extraction of channels and spatial locations. The extracted feature maps are input to the PANet path aggregation network, and the underlying information is stacked with the feature maps. The method improves the detection accuracy of underwater blurred and distorted fish images. Experimental results show that the proposed TAM-YOLO model outperforms YOLOv3, YOLOv4, YOLOv5s, YOLOv5m, and SSD, with a mAP value of 95.88%, thus providing a new strategy for fish detection.

Video stream-based fish feeding behavior recognition has garnered significant attention in recent years, accelerating the optimization of feeding strategies and enhancing aquaculture efficiency. However, current feeding intensity assessment methods suffer from inefficiency and subjectivity in manual observation, compounded by challenges in accurately extracting behavioral features due to high mobility and random movement patterns of outdoor-cultured fish. Constructing an efficient multi-feature extraction model for fish feeding recognition—particularly deployable on mobile and edge devices—remains a critical challenge. To address these limitations, this paper proposes a multi - feature extraction network based on improved MobileViT V3,which uses video streams as input and solves problems of large model size, high computational complexity, and insufficient feature extraction in current models, integrating three key innovations: (1) A Multi-Scale Convolution Module (MSCM) that concurrently captures spatiotemporal, motion, and channel features from video streams; (2) A Feature Fusion Convolutional Block Attention Module (FCBAM) combining shallow-deep features with adaptive attention weighting; (3) A BiasLoss function with dynamic scaling to address intra-class variation and low-quality data. Evaluated on grass carp and crucian carp, our model achieves 97.7% accuracy in feeding intensity classification with only (5.8) M parameters, outperforming C3D-ConvLSTM and MobileNetv3-small baselines while demonstrating enhanced robustness for edge deployment.