The stripe noise effects severely degrade the image quality in infrared imaging systems. The existing destriping algorithms still struggle to balance noise suppression, detail preservation, and real-time performance, which retards their application in spectral imaging and signal processing field. To solve this problem, an innovative wavelet deep neural network from the perspective of transform domain is presented in this paper, which takes the intrinsic characteristics of stripe noise and complementary information between the coefficients of different wavelet sub-bands into full consideration to accurately estimate the noise with the lower computational load. In addition, a special directional regularizer is further defined to separate the scene details from stripe noise more thoroughly and recover the details more accurately. The extensive experiments on simulated and real data demonstrate that our proposed method outperforms several classical destriping methods on both quantitative and qualitative assessments.

To realize the multi-focus image fusion task, an end-to-end deep convolutional neural network (DCNN) model that produces the final fused image directly from the source images is presented in this paper. In order to promote the fusion precision, the innovative multi-focus fusion DCNN introduces a multi-scale feature extraction (MFE) unit to collect more complementary features from different spatial scales and fuse them to excavate more spatial information. Moreover, a visual attention unit is designed to help the network locate the focused region more accurately and pick more useful features for perfectly splicing the details in the fusion process. Experimental results illustrate that the proposed method is superior to several existing multi-focus image fusion methods in both of the subjective visual effects and objective quality metrics.

This paper introduces deformable convolution in deep learning based joint denoising and demosaicing (JDD), which yields more adaptable representation and larger receptive fields in features extraction for a superior restoration performance. However, the deformable convolution generally leads to considerable computational load and irregular memory access bottleneck, limiting its extensive deployment on edge devices. To address this issue, we develop grouping strategy and assign independent offsets to each kernel group to reduce the computation latency while keeping the accuracy. Motivated by the exploration for aggregate distribution characteristics of deformable offsets, we present the offset sharing methodology to simplify the memory access complexity of deformable convolution. As for hardware acceleration, we specially design a novel deformable matrix multiplication workflow incorporated with a deformable memory mapping unit to boost the computational throughput. The verification experiments on FPGA demonstrate that the proposed deformable convolution based JDD can restore 4K Ultra High Definition (UHD) images at 70FPS and yields significant promotion in visual effect and objective quality assessment.

In spite of achieving promising results in hyperspectral image (HSI) restoration, deep-learning-based methodologies still face the problem of spectral or spatial information loss due to neglecting the inner correlation of HSI. To address this issue, we propose an innovative deep recurrent convolution neural network (DnRCNN) model for HSI destriping. To the best of our knowledge, this is the first study on HSI destriping from the perspective of inner band and interband correlation explorations with the recurrent convolution neural network. In the novel DnRCNN, a selective recurrent memory unit (SRMU) is designed to respectively extract the correlative features involved in spectral and spatial domains. Moreover, an innovative recurrent fusion (RF) strategy incorporated with group concatenation is further proposed to remove strip noise and preserve scene details using the complementary features from SRMU. Experimental results on extensive HSI datasets validated that the proposed method achieves a new state-of-the-art (SOTA) HSI destriping performance.

The speckle interference seriously degrades the quality of synthetic aperture radar (SAR) image. The existing despeckling algorithms still struggle to remove noise and preserve details simultaneously. In order to enhance the noise suppression and detail restoration performance, this article specially presents a spatial and transform domain convolutional neural network (STD-CNN) model, which yields an integrated feature representation and learning framework for despeckling. In addition, an innovative feature refinement strategy is proposed to further reduce the detail loss by isolating detail features from noise features. Extensive experiments on synthetic and real SAR images demonstrate that the proposed method outperforms the existing SAR despeckling methods on both quantitative and qualitative assessments. With partial modification, the STD-CNN model can still be extended to other image restoration tasks.