http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Deep Learning for Object Detection Using Micro-Doppler Signatures in Autonomous Vehicles
Godwin Brown Tunze,Jae-Min Lee,Dong-Seong Kim 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.2
This paper proposes a deep learning architecture using micro-Doppler signatures, namely CPNet for autonomous vehicles. Primarily, the network comprises a set of cleverly designed blocks, i.e., CPblocks that learn generic and deep features for efficient extraction of meaningful information from the micro-Doppler signatures. The CPblock deploys two convo-lutional layers with identity mappings, wherein the mappings combine features from the input of the first convolutional to the output of the first convolutional and second convolutional layer via elementwise additional and depthwise concatenation layer, respectively. The configuration of the CPblock decouples the learning of spatial features by using two one-dimensional convolutional layers. Besides, it enhances reuse and new feature exploration via identity mapping. Experimental results indicate that CPNet achieves a detection accuracy of > 96:6%.