通过覆盖率测试理解图像识别中的深度学习模型

View PDF HTML (experimental)

Abstract:Deep neural networks (DNNs) play a crucial role in the field of artificial intelligence, and their security-related testing has been a prominent research focus. By inputting test cases, the behavior of models is examined for anomalies, and coverage metrics are utilized to determine the extent of neurons covered by these test cases. With the widespread application and advancement of DNNs, different types of neural behaviors have garnered attention, leading to the emergence of various coverage metrics for neural networks. However, there is currently a lack of empirical research on these coverage metrics, specifically in analyzing the relationships and patterns between model depth, configuration information, and neural network coverage. This paper aims to investigate the relationships and patterns of four coverage metrics: primary functionality, boundary, hierarchy, and structural coverage. A series of empirical experiments were conducted, selecting LeNet, VGG, and ResNet as different DNN architectures, along with 10 models of varying depths ranging from 5 to 54 layers, to compare and study the relationships between different depths, configuration information, and various neural network coverage metrics. Additionally, an investigation was carried out on the relationships between modified decision/condition coverage and dataset size. Finally, three potential future directions are proposed to further contribute to the security testing of DNN Models.

Subjects:	Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Cite as:	arXiv:2505.08814 [cs.CV]
	(or arXiv:2505.08814v4 [cs.CV] for this version)
	https://doi.org/10.48550/arXiv.2505.08814 arXiv-issued DOI via DataCite

Submission history

From: Wenkai Li [view email]
[v1] Mon, 12 May 2025 08:25:55 UTC (2,751 KB)
[v2] Thu, 15 Jan 2026 14:53:58 UTC (1,903 KB)
[v3] Fri, 3 Jul 2026 12:16:58 UTC (1,903 KB)
[v4] Mon, 12 Oct 2026 06:19:22 UTC (1,902 KB)