Existing road damage detection methods typically rely on a single unified model trained under standard conditions to extract damage features and model spatial relationships between damage regions and the surrounding background. However, road damage exhibits highly diverse relational patterns due to variations in morphology, scale, and contextual appearance, making it difficult for a single model to effectively capture such heterogeneity, especially under complex weather and illumination conditions. To overcome this limitation, we propose a mixture-of-experts–based road damage detection framework that decomposes complex relational modeling into multiple specialized expert processes. A shallow detail-perceptive mixture-of-experts (SDP-MoE) module is introduced to enhance the extraction of fine-grained texture and structural cues critical for accurate damage localization. Meanwhile, a mixture-of-experts gated cross-scale attention (MEGCSA) module is designed to model heterogeneous contextual relationships across multi-scale features, enabling effective integration of local details and global semantics. By collaboratively leveraging specialized experts, the proposed framework provides a flexible and expressive mechanism for multi-scale and multi-type relational modeling in road damage detection. In addition, expert-oriented data augmentation and a load-balancing loss are employed to promote stable and balanced expert learning. Extensive experiments on the CNRDD, RDD2022, and ARSDD benchmarks under challenging low-light and rainy conditions demonstrate that the proposed method consistently outperforms the baseline, achieving m A P @ 0 . 5 improvements of 4.5%, 6.6%, and 4.2%, respectively. More importantly, robustness analysis shows substantially reduced performance degradation in complex road scenarios compared with single-model detectors. Furthermore, the modular and plug-and-play design enables seamless integration into existing road damage detection systems, highlighting its strong practical applicability.

Current single-object tracking algorithms depend on the information supplied by the template to identify and locate the object within the search area. However, environmental complexities and unknown factors can alter the object's state, causing mismatches in template information. The existing works using the template update mechanism (TUM) and multiple template feature fusion have the following problems: 1) TUM is affected by input superposition, making it hard to eliminate noise; 2) they suffer a temporal lag in their responsiveness to changes that occur in the object during the tracking process; 3) it is insufficient to rely solely on visual features within the search area of the current frame to improve the template; and 4) the prior knowledge regarding the input is not fully leveraged to learn the time-variant state of the object. We observe that in complex tracking scenarios, humans subconsciously analyze the evolutionary patterns of the object and its surroundings and integrate this information with the object's initial impression, thereby maintaining an awareness of the object's temporal state. Motivated by this, we propose a novel solution to the above problem, named MtvTrack, which can model the time-variant state of the object through the dynamic evolution pattern and static initial impression. Simultaneously, we propose a method for predicting the evolution pattern of scenes by utilizing past, present, and future (PPF) states. This approach effectively eliminates the information redundancy between consecutive frames and addresses the issue of delayed predictions of the target state in relation to changes within the search area. We establish a joint probability generative model and fully utilize prior knowledge to learn the time-variant state of the object. In addition, we develop a vector quantized-PPF (VQPPF) module for predicting the object's time-variant state. Experimental results on public benchmarks confirm the superior performance of our method. Source code is available at: https://github.com/long-wa/MtvTrack-main.

The advertising budget allocation problem for financial service is dealt with based on statistical learning and evolutionary computation in this paper. Taking the carry-over effects of the advertising into account, the least squares support vector machine regression (LS-SVMR) is used to construct the response model. A comparison between the proposed response model and traditional regression method based market response models is implemented. The results show the effectiveness and validity of the former model. Taking the budgets allocated to every month in the planning horizon as decision variables, the budget allocation optimization model is built and an improved differential evolution algorithm is used to find the optimal solutions. Finally, the proposed budget allocation method is illustrated by a practical problem.