Magnetic particle imaging (MPI) is an emerging medical imaging technique that has high sensitivity, contrast, and excellent depth penetration. In MPI, x-space is a reconstruction method that transforms the measured voltages into particle concentrations. The reconstructed native image can be modeled as a convolution of the magnetic particle concentration with a point-spread function (PSF). The PSF is one of the important parameters in deconvolution. However, accurately measuring or modeling the PSF in the hardware used for deconvolution is challenging due to the various environment and magnetic particle relaxation. The inaccurate PSF estimation may lead to the loss of the content structure of the MPI image, especially in low gradient fields. In this study, we developed a Dual Adversarial Network (DAN) with patch-wise contrastive constraint to deblur the MPI image. This method can overcome the limitations of unpaired data in data acquisition scenarios and remove the blur around the boundary more effectively than the common deconvolution method. We evaluated the performance of the proposed DAN model on simulated and real data. Experimental results confirmed that our model performs favorably against the deconvolution method that is mainly used for deblurring the MPI image and other GAN-based deep learning models.

Ultralow-field (ULF) magnetic resonance imaging (MRI) can suffer from inferior image quality because of low signal-to-noise ratio (SNR). As an efficient way to cover the k-space, the spiral acquisition technique has shown great potential in improving imaging SNR efficiency at ULF. The current study aimed to address the problems of noise and blurring cancelation in the ULF case with spiral trajectory, and we proposed a spiral-out sequence for brain imaging using a portable 50-mT MRI system. The proposed sequence consisted of three modules: noise calibration, field map acquisition, and imaging. In the calibration step, transfer coefficients were obtained between signals from primary and noise-pick-up coils to perform electromagnetic interference (EMI) cancelation. Embedded field map acquisition was performed to correct accumulated phase error due to main field inhomogeneity. Considering imaging SNR, a lower bandwidth for data sampling was adopted in the sequence design because the 50-mT scanner is in a low SNR regime. Image reconstruction proceeded with sampled data by leveraging system imperfections, such as gradient delays and concomitant fields. The proposed method can provide images with higher SNR efficiency compared with its Cartesian counterparts. An improvement in temporal SNR of approximately 23%-44% was measured via phantom and in vivo experiments. Distortion-free images with a noise suppression rate of nearly 80% were obtained by the proposed technique. A comparison was also made with a state-of-the-art EMI cancelation algorithm used in the ULF-MRI system. SNR efficiency-enhanced spiral acquisitions were investigated for ULF-MR scanners and future studies could focus on various image contrasts based on our proposed approach to widen ULF applications.

Underwater target detection and identification technology are currently two of the most important research directions in the information disciplines. Traditionally, underwater target detection technology has struggled to meet the needs of current engineering. However, due to the large manifold error of the underwater sonar array and the complexity of ensuring long-term signal stability, traditional high-resolution array signal processing methods are not ideal for practical underwater applications. In conventional beamforming methods, when the signal-to-noise ratio is lower than -43.05 dB, the general direction can only be vaguely identified in the general direction. To address the above challenges, this paper proposes a beamforming method based on a deep neural network. Through preprocessing, the space-time domain of the target sound signal is converted into two-dimensional data in the angle-time domain. Subsequently, we trained the network with enough sample datasets. Finally, high-resolution recognition and prediction of two-dimensional images are realized. The results of the test dataset in this paper demonstrate the effectiveness of the proposed method, with a minimum signal-to-noise ratio of -48 dB.

Matching infrared (IR) facial probes against a gallery of visible light faces remains a challenge, especially when combined with cross-distance due to deteriorated quality of the IR data. In this paper, we study the scenario where visible light faces are acquired at a short standoff, while IR faces are long-range data. To address the issue of quality imbalance between the heterogeneous imagery, we propose to compensate it by upgrading the lower-quality IR faces. Specifically, this is realized through cascaded face enhancement that combines an existing denoising algorithm (BM3D) with a new deep-learning-based deblurring model we propose (named SVDFace). Different IR bands, short-wave infrared (SWIR) and near-infrared (NIR), as well as different standoffs, are involved in the experiments. Results show that, in all cases, our proposed approach for quality balancing yields improved recognition performance, which is especially effective when involving SWIR images at a longer standoff. Our approach outperforms another easy and straightforward downgrading approach. The cascaded face enhancement structure is also shown to be beneficial and necessary. Finally, inspired by the singular value decomposition (SVD) theory, the proposed deblurring model of SVDFace is succinct, efficient and interpretable in structure. It is proven to be advantageous over traditional deblurring algorithms as well as state-of-the-art deep-learning-based deblurring algorithms.

Multi-spectral imaging (MSI) produces a sequence of spectral images to capture the inner structure of different species, which was recently introduced into ocular disease diagnosis. However, the quality of MSI images can be significantly degraded by motion blur caused by the inevitable saccades and exposure time required for maintaining a sufficiently high signal-to-noise ratio. This degradation may confuse an ophthalmologist, reduce the examination quality, or defeat various image analysis algorithms. We propose an early work specially on deblurring sequential MSI images, which is distinguished from many of the current image deblurring techniques by resolving the blur kernel simultaneously for all the images in an MSI sequence. It is accomplished by incorporating several a priori constraints including the sharpness of the latent clear image, the spatial and temporal smoothness of the blur kernel and the similarity between temporally-neighboring images in MSI sequence. Specifically, we model the similarity between MSI images with mutual information considering the different wavelengths used for capturing different images in MSI sequence. The optimization of the proposed approach is based on a multi-scale framework and stepwise optimization strategy. Experimental results from 22 MSI sequences validate that our approach outperforms several state-of-the-art techniques in natural image deblurring.