CNF: An Automated Test Case Generation Method Based on Neuron Coverage Guidance in Key Network Layers
To address the issue of low efficiency and inability to cover all test scenarios in manually generated test datasets, an automated test case generation method called Critical Network Layer Fuzzy testing (CNF) is proposed, based on the key network layer neuron coverage guidance criterion. The convolu...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10982285/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | To address the issue of low efficiency and inability to cover all test scenarios in manually generated test datasets, an automated test case generation method called Critical Network Layer Fuzzy testing (CNF) is proposed, based on the key network layer neuron coverage guidance criterion. The convolutional layer is selected as the key network layer. It is verified through experiments that the trend of neuron coverage changes in the key network layer is consistent with that of the entire network layer, significantly improving computational efficiency. Then, through visual and quantitative comparisons, the number of affine transformations required for automatic label generation is preliminarily determined. Finally, a network model is trained and comparative experiments are conducted, obtaining final results that verify the effectiveness of the CNF automatic label generation strategy. |
|---|---|
| ISSN: | 2169-3536 |