HIGHLIGHTS
- who: Zhizhan Lu and colleagues from the School of Civil Engineering and Transportation, Beihua University, Jilin, China have published the research: An improved YOLOv5 algorithm for obscured target recognition, in the Journal: (JOURNAL)
- what: For the input incoming feature layer, the authors focus on its weights for each channel. To be able to demonstrate the effectiveness of the optimized network, the paper compares the training times and the actual detection accuracy of the original YOLOv5 network and the YOLOv5 network with the addition of the SE attention mechanism in the same dataset. The authors propose . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.