arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2310.03033cs.CVcs.AIcs.LGcs.LO
收藏

面向增强型交通标志识别的高精度二值神经网络局部鲁棒性基准测试

Benchmarking Local Robustness of High-Accuracy Binary Neural Networks for Enhanced Traffic Sign Recognition

West University of Timisoara(蒂米什瓦拉西部大学)

Andreea Postovan, Mădălina Eraşcu

AI总结:

该研究针对高精度交通标志识别二值神经网络,构建了含多层结构的局部鲁棒性基准测试集,经VNN-COMP'23验证现有求解器的处理能力与问题,并计划探索延长时间的效果及错误结果成因。

摘要:

交通标志在自动驾驶系统的道路安全与交通管理中发挥着关键作用。准确的交通标志分类至关重要,但由于对抗样本、遮挡等现实复杂因素的存在,这一任务颇具挑战性。为解决这些问题,二值神经网络(Binary Neural Networks, BNN)在构建适用于资源受限设备的分类器方面展现出应用前景。\n在我们此前的工作中,我们提出了用于交通标志识别的高精度BNN模型,重点针对有限的计算与能源资源优化模型的紧凑尺寸。为评估这些模型的局部鲁棒性,本文提出了一组基准测试问题,其中包含的网络层对现有最先进的验证工具构成挑战,这些层包括二值化卷积层、最大池化层、批量归一化层、全连接层。验证问题的难度源于以下因素:网络参数量大(90.5万-170万)、输入维度高(2700-12000)、区域数量多(43个),且神经网络不具备稀疏性。\n本文提出的BNN模型与局部鲁棒性属性可在以下网址获取:https://github.com/ChristopherBrix/vnncomp2023_benchmarks/tree/main/benchmarks/traffic_signs_recognition。\n第四届国际神经网络验证竞赛(VNN-COMP'23)的结果显示,7个求解器中有4个能够处理我们随机选取的大量基准测试用例(在45个用例中,最少解决6个,最多解决36个)。出人意料的是,工具还会输出错误结果或缺失反例(数量在1到4个之间)。目前,我们的研究重点是探索通过延长分配时间(此前设定为8分钟)来提升已解决实例数量的可能性。此外,我们还对工具在部分基准测试中产生错误结果的背后原因十分感兴趣。

英文摘要:

Traffic signs play a critical role in road safety and traffic management for autonomous driving systems. Accurate traffic sign classification is essential but challenging due to real-world complexities like adversarial examples and occlusions. To address these issues, binary neural networks offer promise in constructing classifiers suitable for resource-constrained devices. In our previous work, we proposed high-accuracy BNN models for traffic sign recognition, focusing on compact size for limited computation and energy resources. To evaluate their local robustness, this paper introduces a set of benchmark problems featuring layers that challenge state-of-the-art verification tools. These layers include binarized convolutions, max pooling, batch normalization, fully connected. The difficulty of the verification problem is given by the high number of network parameters (905k - 1.7 M), of the input dimension (2.7k-12k), and of the number of regions (43) as well by the fact that the neural networks are not sparse. The proposed BNN models and local robustness properties can be checked at https://github.com/ChristopherBrix/vnncomp2023_benchmarks/tree/main/benchmarks/traffic_signs_recognition. The results of the 4th International Verification of Neural Networks Competition (VNN-COMP'23) revealed the fact that 4, out of 7, solvers can handle many of our benchmarks randomly selected (minimum is 6, maximum is 36, out of 45). Surprisingly, tools output also wrong results or missing counterexample (ranging from 1 to 4). Currently, our focus lies in exploring the possibility of achieving a greater count of solved instances by extending the allotted time (previously set at 8 minutes). Furthermore, we are intrigued by the reasons behind the erroneous outcomes provided by the tools for certain benchmarks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑