Song Liu

Deep learning has rapidly progressed in Synthetic Aperture Radar (SAR) image object detection. However, complex noise, intricate backgrounds, and multiscale objects impact the performance of SAR image object detection. To solve these problems, in this paper, we propose a Feature Separation and Integration Network (FSINet) to mitigate noise interference and improve the accuracy of multiscale objects for SAR images. FSINet is constructed in three steps. First, we construct the Feature Separation Module (FSM) as a preprocessing unit that decouples input images into distinct components, enabling collaboration with the downstream backbone to separate noise and enhance critical information. Subsequently, we propose an Adaptive Feature Integration Module (AFIM) designed to fuse cross-layer multiscale features, generating precise feature representations. Additionally, a new regression loss called Shape Normalized Wasserstein Distance (SNWD) is used to enhance the detection performance for multiscale objects. To validate the superiority of the proposed FSINet, a series of comparative experiments are conducted on three public SAR image object detection datasets, that is the High-Resolution SAR Image Dataset (HRSID), SAR-AIRCraft dataset, and MSAR dataset. The experimental results demonstrate the superiority of our method, revealing that our method achieves state-of-the-art performance in SAR image object detection.

Hyperspectral images (HSIs) are widely used to identify and characterize objects in scenes of interest, but they are associated with high acquisition costs and low spatial resolutions. With the development of deep learning, HSI reconstruction from low-cost and high-spatial-resolution RGB images has attracted widespread attention. It is an inexpensive way to obtain HSIs via the spectral reconstruction (SR) of RGB data. However, due to a lack of consideration of outdoor solar illumination variation in existing reconstruction methods, the accuracy of outdoor SR remains limited. In this paper, we present an attention neural network based on an adaptive weighted attention network (AWAN), which considers outdoor solar illumination variation by prior illumination information being introduced into the network through a basic 2D block. To verify our network, we conduct experiments on our Variational Illumination Hyperspectral (VIHS) dataset, which is composed of natural HSIs and corresponding RGB and illumination data. The raw HSIs are taken on a portable HS camera, and RGB images are resampled directly from the corresponding HSIs, which are not affected by illumination under CIE-1964 Standard Illuminant. Illumination data are acquired with an outdoor illumination measuring device (IMD). Compared to other methods and the reconstructed results not considering solar illumination variation, our reconstruction results have higher accuracy and perform well in similarity evaluations and classifications using supervised and unsupervised methods.

In the field of remote sens., change detection is an important monitoring technology. However, effectively extracting the change feature is still a challenge, especially with an unsupervised method. To solve this problem, we proposed an unsupervised transformer boundary autoencoder network (UTBANet) in this paper. UTBANet consists of a transformer structure and spectral attention in the encoder part. In addition to reconstructing hyperspectral images, UTBANet also adds a decoder branch for reconstructing edge information. The designed encoder module is used to extract features. First, the transformer structure is used for extracting the global features. Then, spectral attention can find important feature maps and reduce feature redundancy. Furthermore, UTBANet reconstructs the hyperspectral image and boundary information simultaneously through two decoders, which can improve the ability of the encoder to extract edge features. Our experiments demonstrate that the proposed structure significantly improves the performance of change detection. Moreover, comparative experiments show that our method is superior to most existing unsupervised methods.

The unmanned aerial vehicle (UAV)-borne hyperspectral imaging system has the advantages of high spatial resolution, flexible operation, under-cloud flying, and easy cooperation with ground synchronous tests. Because this platform often flies under clouds, variations in solar illumination lead to irradiance inconsistency between different rows of hyperspectral images (HSIs). This inconsistency causes errors in radiation correction. In addition, due to the accuracy limitations of the GPS/inertial measurement unit (IMU) and irregular changes in flight platform speed and attitude, HSIs have deformation and drift, which is harmful to the geometric correction and stitching accuracy between flight strips. Consequently, radiation and geometric error limit further applications of large-scale hyperspectral data. To address the above problems, we proposed an integrated solution to acquire and correct UAV-borne hyperspectral images that consist of illumination data acquisition, radiance and geometric correction, HSI, multispectral image (MSI) registration, and multi-strip stitching. We presented an improved three-parameter empirical model based on the illumination correction factor, and it showed that the accuracy of radiation correction considering illumination variation improved, especially in some low signal-to-noise ratio (SNR) bands. In addition, the error of large-scale HSI stitching was controlled within one pixel.

This paper proposed the use of hyperspectral data to classify gastric cancer grading and design of a classifier with a low computational cost.