Recently, the field of vehicle-mounted visual intelligence technology has witnessed a surge of interest in pedestrian detection. Existing algorithms for dense pedestrian detection at intersections face challenges such as high computational weight, complex models that are difficult to deploy, and suboptimal detection accuracy for small targets and highly occluded pedestrians. To address these issues, this paper proposes an improved lightweight multi-scale pedestrian detection algorithm, YOLOv8-CB. The algorithm introduces a lightweight cascade fusion network, CFNet (cascade fusion network), and a CBAM attention module to improve the characterization of multi-scale feature semantics and location information, and it superimposes a bidirectional weighted feature fusion path BIFPN structure to fuse more effective features and improve pedestrian detection performance. It is experimentally verified that compared with the YOLOv8n algorithm, the accuracy of the improved model is increased by 2.4%, the number of model parameters is reduced by 6.45%, and the computational load is reduced by 6.74%. The inference time for a single image is 10.8 ms. The cascade fusion algorithm YOLOv8-CB has higher detection accuracy and is a lighter model for multi-scale pedestrian detection in complex scenes such as streets or intersections. This proposed algorithm presents a valuable approach for device-side pedestrian detection with limited computational resources.

Eddy current inspection has been extensively employed in non-destructive testing of various conductive materials. The coil probe, as a mainstream sensor in the eddy current detection system, inevitably encounters interference from external signals while transmitting its own signal. Therefore, developing techniques to extract valuable signals from noisy ones is crucial for ensuring accurate detection. Carbon fiber composites not only possess significantly lower electrical conductivity compared to conventional metallic materials but also exhibit notable anisotropy. To address this issue, we designed an ‘8’ coil probe set where the excitation coil does not electromagnetically interfere with the detection coil. However, practical applications that require portability and miniaturization pose challenges when utilizing this coil probe set to identify carbon content or defects due to the typically weak output signal. To address this issue, this paper proposes a design that combines the ‘8’ structure of the planar coil probe with the principle of phase-locked amplification to create a dual-phase sensitive phase-locked amplification detection circuit. These specific design ideas were tested using a weak signal, which passed through the preamplifier, secondary amplifier, and band-pass filter comprising the target channel for signal amplification and noise filtering. The effective signal amplitude is proportional to the inverse phase difference between the direct current (DC) signal and inversely proportional to the amplitude of the signal. Finally, the DC signal was passed through an analog-to-digital converter (ADC). The analog-to-digital converter (A/D) is used to collect and calculate the DC signal, enabling the detection of weak electrical signals. Simulation experiments demonstrated that the signal detection circuit has an amplitude error below 0.2% and a phase error below 0.5%. The phase-locked amplification circuit designed in this paper can effectively extract the tiny impedance change signals of the planar coil sensor probe with high sensitivity and good robustness.

In port environments, container stacking at significant heights obstructs satellite signal reception by terminal equipment on container trucks, leading to inaccurate positional tracking data. To address this, it is necessary to predict container truck trajectories to fill in the inaccurate positioning signals. In this study, we collected port container truck trajectory data and compared the predictive performance of long short-term memory (LSTM) networks and transformer models, revealing performance turning points at different prediction steps. Based on these findings, we propose a hybrid model named CT-HybridNet, which integrates the LSTM-based DeepPBM-M model with the transformer-based PatchTST model. Given the independence of the prediction errors of the two models, we assume both errors follow a Gaussian distribution. By performing an affine transformation, the proposed hybrid method’s output also follows a Gaussian distribution. Additionally, an adaptive parameter adjustment mechanism optimises performance, enabling CT-HybridNet to achieve dual improvements in trajectory prediction accuracy and stability, with 15% improvement in short-term accuracy and 20% in long-term performance. This study provides a more accurate and stable technical solution for port container truck trajectory prediction, overcoming issues related to positioning inaccuracies and signal obstructions.