Abstract

The amount of data is vital for the training of deep learning network models. Effective data augmentation methods can significantly improve the training effect of the network model. The paper proposes a data augmentation method of randomly pasting the small target to solve the pin-losing bolt data set problems. After cropping the defined difficult bolt structure and saving it, find the position by uniformly random sampling, and paste it into the original detection image to complete the Paste operation. In the subsequent experiments on the SSD network, the AP of normal bolts on the test data set is 85.22% and the AP of pin-losing bolts is 85.16%. This method expands the pin-losing bolt data set, and lays the foundation for further improving the accuracy of pin-losing bolt detection in the future.

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.