Improved Dense Nested Attention Network Based on Transformer for Infrared Small Target Detection

TL;DR

This paper introduces IDNANet, a novel transformer-based deep learning model that significantly improves infrared small target detection by enhancing feature continuity and addressing class imbalance, validated on new and existing datasets.

Contribution

The paper proposes IDNANet, integrating Swin-transformer and ACmix attention into a dense nested structure, along with a new dataset and loss function for improved infrared small target detection.

Findings

Outperforms state-of-the-art methods in detection probability and false-alarm rate.

Achieves 90.89% mIoU on NUDT-SIRST dataset.

Developed the BIT-SIRST dataset with real and synthetic targets.

Abstract

Infrared small target detection based on deep learning offers unique advantages in separating small targets from complex and dynamic backgrounds. However, the features of infrared small targets gradually weaken as the depth of convolutional neural network (CNN) increases. To address this issue, we propose a novel method for detecting infrared small targets called improved dense nested attention network (IDNANet), which is based on the transformer architecture. We preserve the dense nested structure of dense nested attention network (DNANet) and introduce the Swin-transformer during feature extraction stage to enhance the continuity of features. Furthermore, we integrate the ACmix attention structure into the dense nested structure to enhance the features of intermediate layers. Additionally, we design a weighted dice binary cross-entropy (WD-BCE) loss function to mitigate the negative impact of foreground-background imbalance in the samples. Moreover, we develop a dataset specifically for infrared small targets, called BIT-SIRST. The dataset comprises a significant amount of real-world targets and manually annotated labels, as well as synthetic data and corresponding labels. We have evaluated the effectiveness of our method through experiments conducted on public datasets. In comparison to other state-of-the-art methods, our approach outperforms in terms of probability of detection ($P_d$), false-alarm rate ($F_a$), and mean intersection of union ($mIoU$). The $mIoU$ reaches 90.89\% on the NUDT-SIRST dataset and 79.72\% on the SIRST dataset. The BIT-SIRST dataset and codes are available openly at \href{https://github.com/EdwardBao1006/bit\_sirst}{\color[HTML]{B22222}{https://github.com/EdwardBao1006/bit\_sirst}}.