TY - JOUR
T1 - Rotate-Yolov5 for Aerial Images
AU - Chen, H.
AU - Liu, F. X.
AU - Huang, X. L.
N1 - Publisher Copyright:
© Published under licence by IOP Publishing Ltd.
PY - 2022/6/1
Y1 - 2022/6/1
N2 - In Recent years, great progress has made in object detection. However, since the orientation of object in aerial image is random, the regular horizontal object detection method is not suitable for aerial images. In this paper, we present a Rotate-Yolov5 network based on Yolov5. We use an Adaptive Rotating Anchor Generation Module (ARAGM) to generate anchors with object orientation information. Then the orientation information is used for Rotate-Deformable Convolution Module (R-DCM) to extract features. Finally, we use a decouple detection head as Oriented Object Detection Module (OODM) to yield classification and regression results. Moreover, Rotate-Smooth L1 is used to optimize the loss function. We evaluate the proposed Rotate-Yolov5 on DOTA datasets and the mAP reached 75.4, which demonstrate the superiority of its effectiveness.
AB - In Recent years, great progress has made in object detection. However, since the orientation of object in aerial image is random, the regular horizontal object detection method is not suitable for aerial images. In this paper, we present a Rotate-Yolov5 network based on Yolov5. We use an Adaptive Rotating Anchor Generation Module (ARAGM) to generate anchors with object orientation information. Then the orientation information is used for Rotate-Deformable Convolution Module (R-DCM) to extract features. Finally, we use a decouple detection head as Oriented Object Detection Module (OODM) to yield classification and regression results. Moreover, Rotate-Smooth L1 is used to optimize the loss function. We evaluate the proposed Rotate-Yolov5 on DOTA datasets and the mAP reached 75.4, which demonstrate the superiority of its effectiveness.
UR - http://www.scopus.com/inward/record.url?scp=85132011022&partnerID=8YFLogxK
U2 - 10.1088/1742-6596/2278/1/012038
DO - 10.1088/1742-6596/2278/1/012038
M3 - Conference article
AN - SCOPUS:85132011022
SN - 1742-6588
VL - 2278
JO - Journal of Physics: Conference Series
JF - Journal of Physics: Conference Series
IS - 1
M1 - 012038
T2 - 2022 6th International Conference on Machine Vision and Information Technology, CMVIT 2022
Y2 - 25 February 2022
ER -