Multi-scale feature fusion optical remote sensing target detection method

Abstract

An improved model based on you only look once version 8 (YOLOv8) is proposed to solve the problem of low detection accuracy due to the diversity of object sizes in optical remote sensing images. Firstly, the feature pyramid network (FPN) structure of the original YOLOv8 mode is replaced by the generalized-FPN (GFPN) structure in GiraffeDet to realize the “cross-layer” and “cross-scale” adaptive feature fusion, to enrich the semantic information and spatial information on the feature map to improve the target detection ability of the model. Secondly, a pyramid-pool module of multi atrous spatial pyramid pooling (MASPP) is designed by using the idea of atrous convolution and feature pyramid structure to extract multi-scale features, so as to improve the processing ability of the model for multi-scale objects. The experimental results show that the detection accuracy of the improved YOLOv8 model on DIOR dataset is 92% and mean average precision (mAP) is 87.9%, respectively 3.5% and 1.7% higher than those of the original model. It is proved the detection and classification ability of the proposed model on multi-dimensional optical remote sensing target has been improved.

Keywords

A

References

[1] Liu X B, Liu P, Cai Z H, et al. . Research progress of object detection in optical remote sensing images based on deep learning[J]. Acta automatica sinica. 2021, 47(09): 2078-2089

[2] Huang Y B. Research on remote sensing image object detection method based on deep neural network[D]. 2023, Hangzhou, Hangzhou Dianzi University(in Chinese)

[3] Xu W. Research and implementation of remote sensing image detection technology based on convolutional neural network[D]. 2023, Beijing, Beijing University of Posts and Telecommunications(in Chinese)

[4] Nawaz S A, Li J B, Bhatti U A, et al. . AI-based object detection latest trends in remote sensing, multimedia and agriculture applications[J]. Frontiers in plant science. 2022, 13: 1041514

[5] Zheng J, Hu X B, Wei S Y, et al. . Remote sensing object detection technology based on deep learning[J]. Computer engineering and design. 2019, 45(02): 594-600

[6] Yu M, Cheng M, Jiang H, et al. . Multi-scale object detection in optical remote sensing images using atrous feature pyramid network [C]. The 27th International Conference on Neural Information Processing (ICONIP 2020), November 18–22, 2020, online. 2020, Cham, Springer

[7] Liu S, Zhang L, Lu H C, et al. . Center-boundary dual attention for oriented object detection in remote sensing images[J]. IEEE transactions on geoscience and remote sensing. 2022, 60: 1-14

[8] Yang Y R. Vehicle target detection algorithm based on improved faster R-CNN for remote sensing images[J]. Journal of artificial intelligence practice. 2024, 7(1): 27-33

[9] He D, Shi Q, Liu X, et al. . Generating 2m fine-scale urban tree cover product over 34 metropolises in China based on deep context-aware sub-pixel mapping network[J]. International journal of applied earth observation and geoinformation. 2022, 106: 102667

[10] Ge J, Wang C, Zhang B, et al. . Azimuth-sensitive object detection of high-resolution SAR images in complex scenes by using a spatial orientation attention enhancement network[J]. Remote sensing. 2022, 14(9): 2198

[11] Hu Z, Yang H, Lou T. Dual attention-guided feature pyramid network for instance segmentation of group pigs[J]. Computers and electronics in agriculture. 2021, 186: 106140

[12] Lan J H, Zhang C, Lu W J, et al. . Spatial-transformer and cross-scale fusion network (STCS-Net) for small object detection in remote sensing images[J]. Journal of the Indian society of remote sensing. 2023, 51(7): 1427-1439

[13] Shi W Y, Zhang S W, Zhang S Q. CAW-YOLO: cross-layer fusion and weighted receptive field-based YOLO for small object detection in remote sensing[J]. Computer modeling in engineering & sciences. 2024, 139(3): 3209-3231

[14] Huang W, Li G, Chen Q, et al. . CF2PN: a cross-scale feature fusion pyramid network based remote sensing target detection[J]. Remote sensing. 2021, 13(5): 847

[15] Andrzej S, Goru U K, Garikapati B D, et al. . Spiral search grasshopper features selection with VGG19-ResNet50 for remote sensing object detection[J]. Remote sensing. 2022, 14(21): 5398

[16] Lin T Y, Dollar P, Girshick R, et al. . Feature pyramid networks for object detection [C]. IEEE Conference on Computer Vision and Pattern Recognition, July 21–26, 2017, Hawaii, USA. 2017, New York, IEEE: 2117-2125

[17] Sun H T, Yang S, Chen L J, et al. . Brain tumor image segmentation based on improved FPN[J]. BMC medical imaging. 2023, 23(1): 172

[18] Jiang Y Q, Tan Z Y, Wang J Y. GiraffeDet: a heavy-neck paradigm for detection in remote sensing images using progressive anchor boxes[J]. Remote sensing. 2021, 13(8): 1447

[19] Qiu X. U-Net-ASPP: U-Net based on atrous spatial pyramid pooling model for medical image segmentation in COVID-19[J]. Journal of applied science and engineering. 2022, 25(6): 1167-1176

[20] Park M H, Cho J H, Kim Y T. CNN model with multilayer ASPP and two-step cross-stage for semantic segmentation[J]. Machines. 2023, 11(2): 126

[21] Wang H L, Gui D W, Liu Q, et al. . Vegetation coverage precisely extracting and driving factors analysis in drylands[J]. Ecological informatics. 2024, 79: 102409

RIGHTS & PERMISSIONS

Tianjin University of Technology