Drone detection is a significant research topic due to the potential security threats posed by the misuse of drones in both civilian and military domains. However, traditional drone detection methods are challenged by the drastic scale changes and complex ambiguity during drone flight, and it is difficult to detect small target drones quickly and efficiently. We propose an information-enhanced model based on improved YOLOv5 (TGC-YOLOv5) for fast and accurate detection of small target drones in complex environments. The main contributions of this paper are as follows: First, the Transformer encoder module is incorporated into YOLOv5 to augment attention toward the regions of interest. Second, the Global Attention Mechanism (GAM) is embraced to mitigate information diffusion among distinct layers and amplify the global cross-dimensional interaction features. Finally, the Coordinate Attention Mechanism (CA) is incorporated into the bottleneck part of C3, enhancing the extraction capability of local information for small targets. To enhance and verify the robustness and generalization of the model, a small target drone dataset (SUAV-DATA) is constructed in all-weather, multi-scenario, and complex environments. The experimental results show that based on the SUAV-DATA dataset, the AP value of TGC-YOLOv5 reaches 0.848, which is 2.5% higher than the original YOLOv5, and the Recall value of TGC-YOLOv5 reaches 0.823, which is a 3.8% improvement over the original YOLOv5. The robustness of our proposed model is also verified on the Real-World open-source image dataset, achieving the best accuracy in light, fog, stain, and saturation pollution images. The findings and methods of this paper have important significance and value for improving the efficiency and precision of drone detection.

Artificial intelligence (AI) is transforming electrochemical biosensing systems, offering novel solutions for foodborne pathogen detection. This review examines the integration of AI technologies, particularly machine learning and deep learning algorithms, in enhancing sensor design, material optimization, and signal processing for detecting key pathogens such as Escherichia coli, Salmonella, and Staphylococcus aureus. Key advancements include improved sensitivity, multiplexed detection, and adaptability to complex environments. The application of AI to the design of recognition molecules (e.g., enzymes, antibodies, aptamers), as well as to electrochemical parameter tuning and multicomponent signal analysis, is systematically reviewed. Additionally, the convergence of AI with the Internet of Things (IoT) is discussed as a pathway to portable, real-time detection platforms. The review highlights the pivotal role of AI across multiple layers of biosensor development, emphasizing the opportunities and challenges that arise from interdisciplinary integration and the practical deployment of IoT-enabled technologies in electrochemical sensing systems. Despite significant progress, challenges remain in data quality, model generalization, and interpretability. The review concludes by outlining future research directions for building robust, intelligent biosensing systems capable of supporting scalable food safety monitoring.

An IoT wristband for basketball shooting analysis, which can provide quantitative shooting guidance for basketball players in a convenient, low-cost manner, was developed. A micro inertial measurement unit (IMU) sensor-based wristband was used to collect the shooting motion data of 20 basketball players with different levels of skills: 15 amateur players and 5 elite players. The wristband was 247 mm ${}\times {}20$ mm ${}\times {}11$ mm in size, weighed 12.3 g, and consisted of a power system, a microcontroller, and an IMU sensor (MPU-9250), which can acquire the triaxial data of acceleration, angular velocity, and magnetic field data, and then transmit them to a computing platform via Bluetooth. After correlation analysis, the triaxial data of accelerations and angular velocities were used to identify these 18 shooting movements. Experimental results showed that the four categories of shooting movements (i.e., set shots, layups, jump shots, and tip-ins) could be recognized with an accuracy of 98.0%. The overall recognition accuracy of 18 kinds of shooting movements reached 98.5%. The AI wristband could also be used to distinguish the identity of the participants with a recognition rate of 97.4%. This new technology can be applied to quantitative basketball shot-making guidance and has potential applications for analyzing the motions of other sports such as table tennis, racquetball, volleyball, and soccer.