Multifarious hyperspectral image (HSI) classification methods based on convolutional neural networks (CNN) have been gradually proposed and achieve a promising classification performance. However, hyperspectral image classification still suffers from various challenges, including abundant redundant information, insufficient spectral-spatial representation, irregular class distribution, and so forth. To address these issues, we propose a novel 2D-3D CNN with spectral-spatial multi-scale feature fusion for hyperspectral image classification, which consists of two feature extraction streams, a feature fusion module as well as a classification scheme. First, we employ two diverse backbone modules for feature representation, that is, the spectral feature and the spatial feature extraction streams. The former utilizes a hierarchical feature extraction module to capture multi-scale spectral features, while the latter extracts multi-stage spatial features by introducing a multi-level fusion structure. With these network units, the category attribute information of HSI can be fully excavated. Then, to output more complete and robust information for classification, a multi-scale spectral-spatial-semantic feature fusion module is presented based on a Decomposition-Reconstruction structure. Last of all, we innovate a classification scheme to lift the classification accuracy. Experimental results on three public datasets demonstrate that the proposed method outperforms the state-of-the-art methods.

It is difficult to find correct correspondences for infrared and visible image registration because of different imaging principles. Traditional registration methods based on the point feature require designing the complicated feature descriptor and eliminate mismatched points, which results in unsatisfactory precision and much calculation time. To tackle these problems, this paper presents an artful method based on constrained point features to align infrared and visible images. The proposed method principally contains three steps. First, constrained point features are extracted by employing an object detection algorithm, which avoids constructing the complex feature descriptor and introduces the senior semantic information to improve the registration accuracy. Then, the left value rule (LV-rule) is designed to match constrained points strictly without the deletion of mismatched and redundant points. Finally, the affine transformation matrix is calculated according to matched point pairs. Moreover, this paper presents an evaluation method to automatically estimate registration accuracy. The proposed method is tested on a public dataset. Among all tested infrared-visible image pairs, registration results demonstrate that the proposed framework outperforms five state-of-the-art registration algorithms in terms of accuracy, speed, and robustness.

Infrared and visible image fusion aims to generate an image with prominent target information and abundant texture details. Most existing methods generally rely on manually designing complex fusion rules to realize image fusion. Some deep learning fusion networks tend to ignore the correlation between different level features, which may cause loss of intensity information and texture details in the fused image. To overcome these drawbacks, we propose a multi-level hybrid transmission network for infrared and visible image fusion, which mainly contains the multi-level residual encoder module and the hybrid transmission decoder module. Considering the great difference between infrared and visible images, the multi-level residual encoder module with two independent branches is designed to extract abundant features from source images. To avoid complicated fusion strategies, the concatenate-convolution is applied to fuse features. Towards utilizing information from source images efficiently, the hybrid transmission decoder module is constructed to integrate different level features. Experimental results and analyses on three public datasets demonstrate that our method not only can achieve high quality image fusion but also performs better than comparison methods in terms of qualitative and quantitative comparisons. In addition, the proposed method has good real-time performance in infrared and visible image fusion.

The effectiveness of depth information in saliency detection has been fully proved. However, it is still worth exploring how to utilize the depth information more efficiently. Erroneous depth information may cause detection failure, while non-salient objects may be closer to the camera which also leads to erroneously emphasis on non-salient regions. Moreover, most of the existing RGB-D saliency detection models have poor robustness when the salient object touches the image boundaries. To mitigate these problems, we propose a multi-stage saliency detection model with the bilateral absorbing Markov chain guided by depth information. The proposed model progressively extracts the saliency cues with three level (low-, mid-, and high-level) stages. First, we generate low-level saliency cues by explicitly combining color and depth information. Then, we design a bilateral absorbing Markov chain to calculate mid-level saliency maps. In mid-level, to suppress boundary touch problem, we present the background seed screening mechanism (BSSM) for improving the construction of the two-layer sparse graph and better selecting background-based absorbing nodes. Furthermore, the cross-modal multi-graph learning model (CMLM) is designed to fully explore the intrinsic complementary relationship between color and depth information. Finally, to obtain a more highlighted and homogeneous saliency map in high-level, we structure a depth-guided optimization module which combines cellular automata and suppression-enhancement function pair. This optimization module refines the saliency map in color space and depth space, respectively. Comprehensive experiments on three challenging benchmark datasets demonstrate the effectiveness of our proposed method both qualitatively and quantitatively.

Due to end-to-end optimization characteristics and fine generalization ability, convolutional neural networks have been widely applied to hyperspectral image (HSI) classification, playing an irreplaceable role. However, previous studies struggle with two major challenges: (1) HSI contains complex topographic features, the number of labeled samples in different categories is unbalanced, resulting in poor classification for categories with few labeled samples; (2) With the deepening of neural network models, it is difficult to extract more discriminative spectral-spatial features. To address the issues mentioned above, we propose a discriminative spectral-spatial-semantic feature network based on shuffle and frequency attention mechanisms for HSI classification. There are four main parts of our approach: spectral-spatial shuffle attention module (SSAM), context-aware high-level spectral-spatial feature extraction module (CHSFEM), spectral-spatial frequency attention module (SFAM), and cross-connected semantic feature extraction module (CSFEM). First, to fully excavate the category attribute information, SSAM based on a “Deconstruction-Reconstruction” structure is designed, solving the problem of poor classification performance caused by an unbalanced number of label samples. Considering that deep spectral-spatial features are difficult to extract, CHSFEM and SFAM are constructed. The former is based on the “Horizontal-Vertical” structure to capture context-aware high-level multiscale features. The latter introduces multiple frequency components to compress channels to obtain more multifarious features. Finally, towards suppressing noisy boundaries efficiently and capturing abundant semantic information, CSFEM is devised. Numerous experiments are implemented on four public datasets: the evaluation indexes of OA, AA and Kappa on four datasets all exceed 99%, demonstrating that our method can achieve satisfactory performance and is superior to other contrasting methods.

The depth map contains abundant spatial structure cues, which makes it extensively introduced into saliency detection tasks for improving the detection accuracy. Nevertheless, the acquired depth map is often with uneven quality, due to the interference of depth sensors and external environments, posing a challenge when trying to minimize the disturbances from low-quality depth maps during the fusion process. In this article, to mitigate such issues and highlight the salient objects, we propose a progressive guided fusion network (PGFNet) with multi-modal and multi-scale attention for RGB-D salient object detection. Particularly, we first present a multi-modal and multi-scale attention fusion model (MMAFM) to fully mine and utilize the complementarity of features at different scales and modalities for achieving optimal fusion. Then, to strengthen the semantic expressiveness of the shallow-layer features, we design a multi-modal feature refinement mechanism (MFRM), which exploits the high-level fusion feature to guide the enhancement of the shallow-layer original RGB and depth features before they are fused. Moreover, a residual prediction module (RPM) is applied to further suppress background elements. Our entire network adopts a top-down strategy to progressively excavate and integrate valuable information. Compared with the state-of-the-art methods, experimental results demonstrate the effectiveness of our proposed method both qualitatively and quantitatively on eight challenging benchmark datasets.