TBC-Net: A real-time detector for infrared small target detection using semantic constraint

TL;DR

This paper introduces TBC-Net, a lightweight deep learning model for real-time infrared small target detection, combining semantic constraints and data synthesis to improve accuracy and reduce false alarms in complex backgrounds.

Contribution

The paper presents a novel TBC-Net architecture with a target extraction module and semantic constraint module, along with a joint loss function and data synthesis method for enhanced small target detection.

Findings

TBC-Net outperforms traditional methods in reducing false alarms.

Achieves real-time detection on NVIDIA Jetson AGX Xavier.

Improves small target feature learning with joint loss and semantic constraints.

Abstract

Infrared small target detection is a key technique in infrared search and tracking (IRST) systems. Although deep learning has been widely used in the vision tasks of visible light images recently, it is rarely used in infrared small target detection due to the difficulty in learning small target features. In this paper, we propose a novel lightweight convolutional neural network TBC-Net for infrared small target detection. The TBCNet consists of a target extraction module (TEM) and a semantic constraint module (SCM), which are used to extract small targets from infrared images and to classify the extracted target images during the training, respectively. Meanwhile, we propose a joint loss function and a training method. The SCM imposes a semantic constraint on TEM by combining the high-level classification task and solve the problem of the difficulty to learn features caused by class imbalance problem. During the training, the targets are extracted from the input image and then be classified by SCM. During the inference, only the TEM is used to detect the small targets. We also propose a data synthesis method to generate training data. The experimental results show that compared with the traditional methods, TBC-Net can better reduce the false alarm caused by complicated background, the proposed network structure and joint loss have a significant improvement on small target feature learning. Besides, TBC-Net can achieve real-time detection on the NVIDIA Jetson AGX Xavier development board, which is suitable for applications such as field research with drones equipped with infrared sensors.