Select-Mosaic: Data Augmentation Method for Dense Small Object Scenes

TL;DR

Select-Mosaic is a novel data augmentation technique that improves dense small object detection in aerial images by using a fine-grained region selection strategy, leading to better accuracy and stability.

Contribution

The paper introduces Select-Mosaic, an enhanced mosaic data augmentation method with a region selection strategy tailored for dense small object detection.

Findings

Significantly improves detection accuracy for dense small objects.

Enhances model stability in aerial image detection tasks.

Outperforms traditional mosaic augmentation methods.

Abstract

Data augmentation refers to the process of applying a series of transformations or expansions to original data to generate new samples, thereby increasing the diversity and quantity of the data, effectively improving the performance and robustness of models. As a common data augmentation method, Mosaic data augmentation technique stitches multiple images together to increase the diversity and complexity of training data, thereby reducing the risk of overfitting. Although Mosaic data augmentation achieves excellent results in general detection tasks by stitching images together, it still has certain limitations for specific detection tasks. This paper addresses the challenge of detecting a large number of densely distributed small objects in aerial images by proposing the Select-Mosaic data augmentation method, which is improved with a fine-grained region selection strategy. The improved Select-Mosaic method demonstrates superior performance in handling dense small object detection tasks, significantly enhancing the accuracy and stability of detection models. Code is available at https://github.com/malagoutou/Select-Mosaic.