Wupeng Deng

As unmanned aerial vehicles (UAVs) have increasingly become aerial imaging platforms, the reliance of traditional photogrammetry on ground control points (GCPs) remains a major limitation in complex terrain, confined spaces, and scenarios where control points are difficult to deploy. To address this issue, this study proposes a proof-of-concept framework for free-flight photogrammetry based on the fusion of monocular vision and sensor-group displacement information. The framework employs a rigid point set station-displacement algorithm to compute the exterior orientation elements between adjacent measurement stations, providing a feasible approach for multi-station pose propagation under control-point-free conditions. In addition, a composite weighting strategy incorporating the effects of optical distortion and rigid-body consistency evaluation is developed to improve the rational use of point-set information during station-displacement computation. To evaluate the feasibility of the proposed method, numerical simulations were first conducted to analyze the variation patterns of exterior orientation computation and target-point reconstruction under different sampling intervals and error conditions. Subsequently, an indoor controlled bench-top experimental platform was constructed to physically validate the complete workflow of the proposed method. The bench-top experimental results show that the overall mean three-dimensional positioning error of the two cross-station image pairs was 15.450 mm, and the maximum three-dimensional positioning error was 36.685 mm. The mean absolute distance errors for station 1–station 2 and station 1–station 3 were 9.230 mm and 12.436 mm, respectively. These results indicate that the proposed method can complete station-displacement-based exterior orientation computation and three-dimensional target measurement in a controlled physical scenario, demonstrating clear proof-of-concept significance. It should be noted that UAV measurement experiments under real flight conditions have not yet been completed in this study, and further validation on an actual UAV platform is still required.

In recirculating aquaculture systems, fish detection is an essential component for maintaining effective farming operations. The availability of high-quality fish datasets is limited because of the richness of fish species, and the annotation of large-scale data, which is used to train models, is often labor-intensive and time-consuming. The presence of different fish species across batches introduces further challenges for consistent detection performance. This work introduces a few-shot learning approach for fish detection, utilizing a customized dataset as novel classes and the Fish4Knowledge dataset for base classes, thereby establishing a framework that enhances adaptability in data-scarce scenarios. Within the model architecture, multi-scale feature extraction is enhanced through an attention mechanism, which is integrated as a dedicated module to strengthen representation learning, thus enhancing the model’s capability to differentiate visually similar fish species. Two distinct customized fish datasets are employed to evaluate the robustness of the proposed method. Experimental results show that the proposed model performs competitively against TFA, Meta-RCNN, and VFA. In the base-training phase, it achieves a mAP of 0.775, slightly surpassing VFA, while in the 1-shot, 5-shot, and 10-shot fine-tuning settings, it obtains mAP values of 0.152, 0.247, and 0.265, respectively. A similar trend is observed on a subset of black fish, with mAP scores of 0.169, 0.253, and 0.286 in the corresponding few-shot settings. These results indicate that the proposed approach can maintain relatively stable detection accuracy and adaptability across different fish batches, offering a practical solution for fish detection tasks in aquaculture when annotated data is scarce. To further demonstrate the efficacy and practical utility of the proposed methodology, a case study in fish farming confirms that the enhanced model achieves consistent and precise detection across diverse fish species, even when trained with limited annotated data.

Fishways can effectively validate the effectiveness and rationality of their construction, optimize operational modes, and achieve intelligent scientific management through fish species detection. Traditional fish species detection methods for fishways are unsuitable due to inefficiency and disruption of the fish ecological environment. Therefore, combining cameras with target detection technology provides a better solution. However, challenges include the limited computational power of onsite equipment, the complexity of model deployment, low detection accuracy, and slow detection speed, all of which are significant obstacles. This paper proposes a fish detection model for accurate and efficient fish detection. Firstly, the backbone network integrates FasterNet-Block, C2f, and an efficient multi-scale EMA attention mechanism to address attention dispersion problems during feature extraction, delivering real-time object detection across different scales. Secondly, the Neck introduces a novel architecture to enhance feature fusion by integrating the RepBlock and BiFusion modules. Finally, the performance of the fish detection model is demonstrated based on the Fish26 dataset, in which the detection accuracy, computational cost, and parameter count are significantly optimized by 1.7%, 23.4%, and 24%, respectively, compared to the state-of-the-art model. At the same time, we installed detection devices in a specific fishway and deployed the proposed method within these devices. We collected data on four fish species passing through the fishway to create a dataset and train the model. The results of the practical application demonstrated superior fish detection capabilities, with rapid detection ability achieved while minimizing resource usage. This validated the effectiveness of the proposed method for equipment deployment in real-world engineering environments. This marks a shift from traditional manual detection to intelligent fish species detection in fishways, promoting water resource utilization and the protection of fish ecological environments.

Disassembly is a key step in remanufacturing, especially for end-of-life (EoL) products such as electric vehicle (EV) batteries, which are challenging to dismantle due to uncertainties in their condition and potential risks of fire, fumes, explosions, and electrical shock. To address these challenges, this paper presents a robotic teleoperation system that leverages augmented reality (AR) and digital twin (DT) technologies to enable a human operator to work away from the danger zone. By integrating AR and DTs, the system not only provides a real-time visual representation of the robot’s status but also enables remote control via gesture recognition. A bidirectional communication framework established within the system synchronises the virtual robot with its physical counterpart in an AR environment, which enhances the operator’s understanding of both the robot and task statuses. In the event of anomalies, the operator can interact with the virtual robot through intuitive gestures based on information displayed on the AR interface, thereby improving decision-making efficiency and operational safety. The application of this system is demonstrated through a case study involving the disassembly of a busbar from an EoL EV battery. Furthermore, the performance of the system in terms of task completion time and operator workload was evaluated and compared with that of AR-based control methods without informational cues and ‘smartpad’ controls. The findings indicate that the proposed system reduces operation time and enhances user experience, delivering its broad application potential in complex industrial settings.