A Survey of Small Object Detection Based on Deep Learning in Aerial Images

Abstract

Small object detection poses a formidable challenge in the field of computer vision, particularly when it comes to analyzing aerial remote sensing images. Despite the rapid development of deep learning and significant progress in detection techniques in natural scenes, the migration of these algorithms to aerial images has not met expectations. This is primarily due to limitations in imaging acquisition conditions, including small target size, viewpoint specificity, background complexity, as well as scale and orientation diversity. Although the increasing application of deep learning-based algorithms to overcome these problems, few studies have summarized the optimization of different deep learning strategies used for small target detection in aerial images. Therefore, this paper aims to explore the application of deep learning methods for small object detection in aerial images. The primary challenges in small object detection in aerial images will be summarized. Next, a meticulous analysis and categorization of the prevailing deep learning optimization strategies employed to surmount the challenges encountered in aerial image detection is undertaken. Following that, we provide a comprehensive presentation of the object detection datasets utilized in aerial remote sensing images, along with the evaluation metrics employed. Additionally, we furnish experimental data pertaining to the currently proposed detection algorithms. Finally, the advantages and disadvantages of various optimization strategies and potential development trends are discussed. Hopefully, it can provide a reference for researchers in this field.

Multidomain feature fusion method for small object classification: MDFF

“…Therefore, the use of the MDFF method can increase the information contained in the proposals, which is highly beneficial for generating high-quality proposals. To demonstrate the feasibility of this viewpoint, we made a simple modification to the ROI-HEAD of faster R-CNN 32 by adding a frequency domain branch for class prediction and named it MDFF-faster R-CNN 33 . We trained it on a subset of the COCO 34 dataset (named person-car).…”

“…Visualization of detection results. (a)–(c) The original image of person-car (a subset of COCO 34 ), faster R-CNN 32 detection results, and MDFF-faster R-CNN 33 detection results, respectively. (d)–(f) The original image of UAV-human, 35 faster R-CNN detection results, and MDFF-faster R-CNN detection results, respectively.…”

Defect detection method of lithium battery based on improved YOLOv7

“…Enter feature diagram F. The channel attention module produces a one-dimensional channel attention map MC., The spatial attention module generates a two-dimensional spatial attention graph MS. Multiply F with MC to get the channel attention feature diagram F ',Then multiply F 'with MS to get the output feature diagram F" 15 ,The calculation formula is as follows…”