The development of object detection in infrared images has attracted more attention in recent years. However, there are few studies on multi-scale object detection in infrared street scene images. Additionally, the lack of high-quality infrared datasets hinders research into such algorithms. In order to solve these issues, we firstly make a series of modifications based on Faster Region-Convolutional Neural Network (R-CNN). In this paper, a double-layer region proposal network (RPN) is proposed to predict proposals of different scales on both fine and coarse feature maps. Secondly, a multi-scale pooling module is introduced into the backbone of the network to explore the response of objects on different scales. Furthermore, the inception4 module and the position sensitive region of interest (ROI) align (PSalign) pooling layer are utilized to explore richer features of the objects. Thirdly, this paper proposes instance level data augmentation, which takes into account the imbalance between categories while enlarging dataset. In the training stage, the online hard example mining method is utilized to further improve the robustness of the algorithm in complex environments. The experimental results show that, compared with baseline, our detection method has state-of-the-art performance.

There are many algorithms that can be used to fuse sensor data. The complementary filtering algorithm has low computational complexity and good real-time performance characteristics. It is very suitable for attitude estimation of small unmanned aerial vehicles (micro-UAVs) equipped with low-cost inertial measurement units (IMUs). However, its low attitude estimation accuracy severely limits its applications. Though, many methods have been proposed by researchers to improve attitude estimation accuracy of complementary filtering algorithms, there are few studies that aim to improve it from the data processing aspect. In this paper, a real-time first-order differential data processing algorithm is proposed for gyroscope data, and an adaptive adjustment strategy is designed for the parameters in the algorithm. Besides, the differential-nonlinear complementary filtering (D-NCF) algorithm is proposed by combine the first-order differential data processing algorithm with the basic nonlinear complementary filtering (NCF) algorithm. The experimental results show that the first-order differential data processing algorithm can effectively correct the gyroscope data, and the Root Mean Square Error (RMSE) of attitude estimation of the D-NCF algorithm is smaller than when the NCF algorithm is used. The RMSE of the roll angle decreases from 1.1653 to 0.5093, that of the pitch angle decreases from 2.9638 to 1.5542, and that of the yaw angle decreases from 0.9398 to 0.6827. In general, the attitude estimation accuracy of D-NCF algorithm is higher than that of the NCF algorithm.

Accurately perceiving location and scene is crucial for autonomous driving and mobile robots. Recent advances in deep learning have made it possible to learn egomotion and depth from monocular images in a self-supervised manner, without requiring highly precise labels to train the networks. However, monocular vision methods suffer from a limitation known as scale-ambiguity, which restricts their application when absolute-scale is necessary. To address this, we propose SelfOdom, a self-supervised dual-network framework that can robustly and consistently learn and generate pose and depth estimates in global scale from monocular images. In particular, we introduce a novel coarse-to-fine training strategy that enables the metric scale to be recovered in a two-stage process. Furthermore, SelfOdom is flexible and can incorporate inertial data with images, which improves its robustness in challenging scenarios, using an attention-based fusion module. Our model excels in both normal and challenging lighting conditions, including difficult night scenes. Extensive experiments on public datasets have demonstrated that SelfOdom outperforms representative traditional and learning-based VO and VIO models.