Abstract: Addressing the issue of inter-class similarity is a challenging task in the research of autonomous driving scene classification, which primarily focuses on learning the distinctive features of targets in real-world complex traffic scenarios with high similarity, and constructing the overall correlation between features for scene classification. To this end, a multi-scale adaptive feature selection network for autonomous driving scene classification is proposed. Initially, a dual multi-scale feature extraction module is utilized for preliminary processing to extract inter-class similar features at different scales. Subsequently, a feature differentiation screening module is designed to complete the screening of scene-similar features, enabling the network to focus more on the typical and easily distinguishable features of different scene categories. Then, the multi-scale feature maps are conveyed together to the feature fusion classification module for scene classification, capturing the correlation between scene features. Finally, an adaptive learning algorithm dynamically adjusts the training parameters through the output results, accelerating the network's convergence speed and improving accuracy. The proposed method is compared with existing network methods on three datasets: BDD100k, BDD100k+, and self-made dataset. Compared with the Top2 networks, it leads in accuracy by 3.29%, 5.59%, and 12.65% (relatively), respectively. The experimental results demonstrate the effectiveness of the proposed method and its strong generalization capability. The scene classification method presented in this paper aims to learn the typical and easily distinguishable features and their correlations under different complex scene categories, reducing the impact of inter-class similarity among multiple targets, thereby making the scene classification results in real-world traffic scenario datasets more accurate.

Key words: autonomous driving, scene understanding, interclass similarity, multi-scale structure, feature screening, adaptive training

摘要： 解决类间相似度问题是自动驾驶场景分类研究中一项充满挑战的任务，主要研究在相似度较高的真实复杂交通场景中，利用网络学习目标特征的差异性，并构建特征之间整体关联性进行场景分类。为此，提出一种多尺度自适应特征筛选的自动驾驶场景分类网络。首先，采用双重多尺度特征提取模块预处理，初步提取不同尺度下的类间相似特征；其次，设计了特征分化筛选模块完成场景相似特征筛选，使网络更关注不同场景类别的典型易区分特征；然后，和多尺度特征图共同传递至特征融合分类模块进行场景分类，捕捉场景特征之间的关联性；最后，由自适应学习算法通过输出结果动态调整训练参数，加快网络收敛速度并提升精度。所提方法在三种数据集BDD100k、BDD100k+和自制数据集上与现有网络方法进行比较，相较Top2网络在精度上分别领先了3.29%、5.59%、12.65%（相对），实验结果表明了所提方法的有效性，并展现了很好的泛化能力。本文提出的场景分类方法旨在学习不同复杂场景类别下的典型易区分的特征及其关联性，降低多目标类间相似的影响，使得在真实交通场景数据集中场景分类结果更加准确。

关键词: 自动驾驶, 场景分类, 类间相似, 多尺度结构, 特征筛选, 自适应训练