Salient Regions Research Articles

Registration of Large Optical and SAR Images with Non-Flat Terrain by Investigating Reliable Sparse Correspondences

Optical and SAR image registration is the primary procedure to exploit the complementary information from the two different image modal types. Although extensive research has been conducted to narrow down the vast radiometric and geometric gaps so as to extract homogeneous characters for feature point matching, few works have considered the registration issue for non-flat terrains, which will bring in more difficulties for not only sparse feature point matching but also outlier removal and geometric relationship estimation. This article addresses these issues with a novel and effective optical-SAR image registration framework. Firstly, sparse feature points are detected based on the phase congruency moment map of the textureless SAR image (SAR-PC-Moment), which helps to identify salient local regions. Then a template matching process using very large local image patches is conducted, which increases the matching accuracy by a significant margin. Secondly, a mutual verification-based initial outlier removal method is proposed, which takes advantage of the different mechanisms of sparse and dense matching and requires no geometric consistency assumption within the inliers. These two procedures will produce a putative correspondence feature point (CP) set with a low outlier ratio and high reliability. In the third step, the putative CPs are used to segment the large input image of non-flat terrain into dozens of locally flat areas using a recursive random sample consensus (RANSAC) method, with each locally flat area co-registered using an affine transformation. As for the mountainous areas with sharp elevation variations, anchor CPs are first identified, and then optical flow-based pixelwise dense matching is conducted. In the experimental section, ablation studies using four precisely co-registered optical-SAR image pairs of flat terrain quantitatively verify the effectiveness of the proposed SAR-PC-Moment-based feature point detector, big template matching strategy, and mutual verification-based outlier removal method. Registration results on four 1 m-resolution non-flat image pairs prove that the proposed framework is able to produce robust and quite accurate registration results.

Optic disc detection based on fully convolutional network and weighted matrix recovery model.

Eye diseases often affect human health. Accurate detection of the optic disc contour is one of the important steps in diagnosing and treating eye diseases. However, the structure of fundus images is complex, and the optic disc region is often disturbed by blood vessels. Considering that the optic disc is usually a saliency region in fundus images, we propose a weakly-supervised optic disc detection method based on the fully convolution neural network (FCN) combined with the weighted low-rank matrix recovery model (WLRR). Firstly, we extract the low-level features of the fundus image and cluster the pixels using the Simple Linear Iterative Clustering (SLIC) algorithm to generate the feature matrix. Secondly, the top-down semantic prior information provided by FCN and bottom-up background prior information of the optic disc region are used to jointly construct the prior information weighting matrix, which more accurately guides the decomposition of the feature matrix into a sparse matrix representing the optic disc and a low-rank matrix representing the background. Experimental results on the DRISHTI-GS dataset and IDRiD dataset show that our method can segment the optic disc region accurately, and its performance is better than existing weakly-supervised optic disc segmentation methods. Graphical abstract of optic disc segmentation.

Retraction Note: Recurrent learning of context for salient region detection

Retraction Note: Recurrent learning of context for salient region detection

360° Image Saliency Prediction by Embedding Self-Supervised Proxy Task

The development of Metaverse industry produces many 360 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$^{\circ}$</tex-math> </inline-formula> images and videos. Transmitting these images or videos efficiently is the key to success of Metaverse. Since the subject’s field of view is limited in Metaverse, from the perception perspective, bit rates can be saved by focusing video encoding on salient regions. On different ways of handling 360 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$^{\circ}$</tex-math> </inline-formula> image projections, the existing works either consider combining local and global projections or just use only global projection for saliency prediction, which results in slow detection speed or low accuracy. In this work, we address this problem by Embedding a self-supervised Proxy task in the Saliency prediction Network, dubbed as <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">EPSNet</b> . The main architecture follows an autoencoder with an encoder for feature extraction and a decoder for saliency prediction. The proxy task is combined with the encoder to enforce it to learn local and global information. It is designed to find the location of a certain local projection in the global projection via self-supervised learning. A cross-attention fusion mechanism is used to fuse the global and local features for location prediction. Then, the decoder is trained based on the sole global projection. In this way, the time-consuming local-global feature fusion is placed in the training stage only. Experiments on public dataset show that our method has achieved satisfactory results in terms of inference speed and accuracy. The dataset and code are available at https://github.com/zzz0326/EPSNet.

Visual Place Recognition via a Multitask Learning Method With Attentive Feature Aggregation

Visual place recognition has gained popularity in recent years. Mainstream convolutional neural network-based methods formulate it as a ranking task and optimize it in the paradigm of deep metric learning, however, the ranking-motivated losses concern only the ranking relationship for each query image and the compactness of intra-place feature distribution is ignored. It is still challenging due to varying viewpoints, illuminations and even dynamic objects. In this paper, a novel multi-task learning framework is proposed, which combines the existing triplet ranking task and our designed binary classification task to jointly optimize the network for better generalization capability. Specifically, a binary classification network with the corresponding binary cross-entropy loss is designed in the classification task. In this way, the intra-place feature compactness and inter-place feature separability are reinforced. At testing stage, this classification network is discarded without increasing the computation cost. Furthermore, an attention module is presented to promote the network to concentrate on the salient regions by assigning different importance to each spatial position. Our method achieves the top-10 recalls of 97.27%, 94.6%, and 96.93% on Pitts250k-test, Tokyo 24/7, and TokyoTM-val datasets, respectively. Extensive experiments prove that the proposed network can learn discriminative global features with better robustness to viewpoints and environmental variations.

SAR ship detection based on salience region extraction and multi-branch attention

SAR ship detection based on salience region extraction and multi-branch attention

A novel seminar learning framework for weakly supervised salient object detection

Weakly supervised salient object detection (SOD) is a challenging task and has drawn much attention from several research perspectives, it has revealed two problems while driving the rapid development of saliency detection. (1) Large divergence in the characteristics of saliency regions in terms of location, shape and size makes them difficult to recognize. (2) The properties of convolutional neural networks dictate that it is insensitive to various transformations, which will lead to hardly balance the application of various disturbances. To tackle these limitations, this paper proposes a novel seminar learning framework with consistent transformation ensembling (SLF-CT) for scribble supervised SOD. The framework consists of the teacher–student model and the student–student model for segmenting the salient objects. Specifically, we first design a cross attention guided network (CAGNet) as a baseline model for saliency prediction. Then we assign CAGNet to the teacher–student model, where the teacher network is based on the exponential moving average and guides the training of the student network. Moreover, we adopt multiple pseudo labels to transfer the information among students from different conditions. To further enhance the regularization of the network, a consistency transformation mechanism is also incorporated, which encourages the saliency prediction and input image of the network to be consistent. The experimental results demonstrate that the proposed approach performs favorably comparable with the state-of-the-art weakly supervised methods. As far as we know, the proposed approach is the first application of seminar learning in the SOD area.

Brain Tumor Detection Based on Deep Learning Approaches and Magnetic Resonance Imaging.

The rapid development of abnormal brain cells that characterizes a brain tumor is a major health risk for adults since it can cause severe impairment of organ function and even death. These tumors come in a wide variety of sizes, textures, and locations. When trying to locate cancerous tumors, magnetic resonance imaging (MRI) is a crucial tool. However, detecting brain tumors manually is a difficult and time-consuming activity that might lead to inaccuracies. In order to solve this, we provide a refined You Only Look Once version 7 (YOLOv7) model for the accurate detection of meningioma, glioma, and pituitary gland tumors within an improved detection of brain tumors system. The visual representation of the MRI scans is enhanced by the use of image enhancement methods that apply different filters to the original pictures. To further improve the training of our proposed model, we apply data augmentation techniques to the openly accessible brain tumor dataset. The curated data include a wide variety of cases, such as 2548 images of gliomas, 2658 images of pituitary, 2582 images of meningioma, and 2500 images of non-tumors. We included the Convolutional Block Attention Module (CBAM) attention mechanism into YOLOv7 to further enhance its feature extraction capabilities, allowing for better emphasis on salient regions linked with brain malignancies. To further improve the model's sensitivity, we have added a Spatial Pyramid Pooling Fast+ (SPPF+) layer to the network's core infrastructure. YOLOv7 now includes decoupled heads, which allow it to efficiently glean useful insights from a wide variety of data. In addition, a Bi-directional Feature Pyramid Network (BiFPN) is used to speed up multi-scale feature fusion and to better collect features associated with tumors. The outcomes verify the efficiency of our suggested method, which achieves a higher overall accuracy in tumor detection than previous state-of-the-art models. As a result, this framework has a lot of potential as a helpful decision-making tool for experts in the field of diagnosing brain tumors.

Global Guided Cross-Modal Cross-Scale Network for RGB-D Salient Object Detection.

RGB-D saliency detection aims to accurately localize salient regions using the complementary information of a depth map. Global contexts carried by the deep layer are key to salient objection detection, but they are diluted when transferred to shallower layers. Besides, depth maps may contain misleading information due to the depth sensors. To tackle these issues, in this paper, we propose a new cross-modal cross-scale network for RGB-D salient object detection, where the global context information provides global guidance to boost performance in complex scenarios. First, we introduce a global guided cross-modal and cross-scale module named G2CMCSM to realize global guided cross-modal cross-scale fusion. Then, we employ feature refinement modules for progressive refinement in a coarse-to-fine manner. In addition, we adopt a hybrid loss function to supervise the training of G2CMCSNet over different scales. With all these modules working together, G2CMCSNet effectively enhances both salient object details and salient object localization. Extensive experiments on challenging benchmark datasets demonstrate that our G2CMCSNet outperforms existing state-of-the-art methods.

Multiscale Feature Aggregation Network for Salient Object Detection in Optical Remote Sensing Images

Optical remote sensing images (ORSI) have various applications in different fields, and salient target detection (ORSI-SOD) of ORSI has become an important research topic in recent years. However, ORSI-SOD is a challenging problem due to the variable and complex backgrounds, large differences in levels, mixed backgrounds, and diverse topological shapes of ORSI. In this paper, we propose a novel model called multi-scale feature refinement aggregation network (MFANet), which consists of a multi-scale feature refinement module (MFR) and a context aggregation module (CFA). The MFR module extracts semantic information of ORSI across different dimensions in the multi-scale feature extraction stage. In the feature refinement stage, we use the proposed self-refinement module under the guidance of attention and reverse attention to progressively refine the prediction results. The CFA module introduces the hybrid attention module to gradually aggregate and extract salient regions from the context extraction module. To adapt to dense scenes, we develop a hybrid loss function that enables network optimization of multi-scale objectives in a self-adaptive manner. Our method outperforms most state-of-the-art salient object detection methods proposed in recent years in terms of accuracy.

Superpixel-based adaptive salient region analysis for infrared and visible image fusion

Superpixel-based adaptive salient region analysis for infrared and visible image fusion

A Salient Object Detection Method Based on Boundary Enhancement.

Visual saliency refers to the human's ability to quickly focus on important parts of their visual field, which is a crucial aspect of image processing, particularly in fields like medical imaging and robotics. Understanding and simulating this mechanism is crucial for solving complex visual problems. In this paper, we propose a salient object detection method based on boundary enhancement, which is applicable to both 2D and 3D sensors data. To address the problem of large-scale variation of salient objects, our method introduces a multi-level feature aggregation module that enhances the expressive ability of fixed-resolution features by utilizing adjacent features to complement each other. Additionally, we propose a multi-scale information extraction module to capture local contextual information at different scales for back-propagated level-by-level features, which allows for better measurement of the composition of the feature map after back-fusion. To tackle the low confidence issue of boundary pixels, we also introduce a boundary extraction module to extract the boundary information of salient regions. This information is then fused with salient target information to further refine the saliency prediction results. During the training process, our method uses a mixed loss function to constrain the model training from two levels: pixels and images. The experimental results demonstrate that our salient target detection method based on boundary enhancement shows good detection effects on targets of different scales, multi-targets, linear targets, and targets in complex scenes. We compare our method with the best method in four conventional datasets and achieve an average improvement of 6.2% on the mean absolute error (MAE) indicators. Overall, our approach shows promise for improving the accuracy and efficiency of salient object detection in a variety of settings, including those involving 2D/3D semantic analysis and reconstruction/inpainting of image/video/point cloud data.

CLT-YOLOX: Improved YOLOX Based on Cross-Layer Transformer for Object Detection Method Regarding Insect Pest

This paper presents an enhanced YOLOX-based algorithm for pest detection, adopting a nature-inspired approach for refining its methodology. To tackle the limited availability of image data pertaining to pests and diseases, the paper incorporates Mosaic and Mixup technologies for effective image preprocessing. Furthermore, a novel training strategy is proposed to enhance the overall quality of the results. The existing architecture is enriched by integrating shallow information, while the CLT module is devised to facilitate cross-layer fusion and extract essential feature information. This advancement enables improved object detection across various scales. Additionally, the paper optimizes the original PFPN structure by eliminating the convolutional layer preceding upsampling, enhancing the C3 module, and integrating the convolutional attention model (CBAM) to identify salient regions within complex scenes. The performance of the proposed CLT-YOLOX model is extensively evaluated using the IP102 dataset, demonstrating its effectiveness. Notably, the model exhibits significant improvements compared to the original AP evaluation index, with an increase of 2.2% in average precision (mAP) and 1.8% in AP75. Furthermore, favorable results are achieved in the COCOmAP index, particularly in the APsmall category where there is a 2.2% improvement in performance.

Clinical Interpretability of Deep Learning for Predicting Microvascular Invasion in Hepatocellular Carcinoma by Using Attention Mechanism.

Preoperative prediction of microvascular invasion (MVI) is essential for management decision in hepatocellular carcinoma (HCC). Deep learning-based prediction models of MVI are numerous but lack clinical interpretation due to their "black-box" nature. Consequently, we aimed to use an attention-guided feature fusion network, including intra- and inter-attention modules, to solve this problem. This retrospective study recruited 210 HCC patients who underwent gadoxetate-enhanced MRI examination before surgery. The MRIs on pre-contrast, arterial, portal, and hepatobiliary phases (hepatobiliary phase: HBP) were used to develop single-phase and multi-phase models. Attention weights provided by attention modules were used to obtain visual explanations of predictive decisions. The four-phase fusion model achieved the highest area under the curve (AUC) of 0.92 (95% CI: 0.84-1.00), and the other models proposed AUCs of 0.75-0.91. Attention heatmaps of collaborative-attention layers revealed that tumor margins in all phases and peritumoral areas in the arterial phase and HBP were salient regions for MVI prediction. Heatmaps of weights in fully connected layers showed that the HBP contributed the most to MVI prediction. Our study firstly implemented self-attention and collaborative-attention to reveal the relationship between deep features and MVI, improving the clinical interpretation of prediction models. The clinical interpretability offers radiologists and clinicians more confidence to apply deep learning models in clinical practice, helping HCC patients formulate personalized therapies.

Spatial frequency enhanced salient object detection

Many salient object detection (SOD) methods based on convolutional neural networks utilize spatial frequency information from images to obtain salient features. However, most of these methods do not extract and/or exploit spatial frequency information effectively. This results in suboptimal detection performance in complex natural and human social scenes. In this work, we first propose an effective and flexible spatial frequency enhancement (SFE) module based on generalized Oct-convolution. SFE can effectively extract and incorporate multiple spatial frequency information from in-stage and cross-stage feature maps and output an arbitrary number of comprehensive and compact frequency features. Then we design a spatial frequency enhanced network (SFENet) adopting two SFE modules to refine high- and low-frequency salient features end-to-end. High-and low-frequency salient features are then integrated into a final full-band saliency prediction. Experiments show that SFENet can ensure the integrity and sharpness of salient regions in complex backgrounds on five benchmark datasets.

Mask-guided network for image captioning

Attention mechanisms have been widely adopted for image captioning because of their powerful performance. In this paper, we propose a Mask Captioning Network (MaC) consisting of an object layer and a background layer to capture the objects and scenes of an image to generate a sentence. To this end, we leverage the Mask RCNN to detect salient regions at the pixel level instead of a bounding box in the object layer. Meanwhile, in the background layer, a CNN model is used to encode the scene features. In addition, MaC is implemented in both LSTM-based and Transformer-based image captioning architectures. We introduce a mask-guided transformer encoder with additional features to enhance the model. Experimental results show that our model significantly outperforms (with a much richer sentence) baseline models and achieves comparable results with state-of-the-art methods on MSCOCO and Flickr30k datasets.

Cross-Modality Double Bidirectional Interaction and Fusion Network for RGB-T Salient Object Detection

RGB-T salient object detection (SOD) aims to detect and segment saliency regions on RGB images and the corresponding thermal maps. The ability of alleviating the modality difference between RGB and thermal modality plays a vital role in the development of RGB-T SOD. However, most of the existing methods try to integrate multi-modal information through various fusion strategies, or reduce the modality difference via unidirectional or undifferentiated bidirectional interaction, but failing in some challenging scenes. To deal with the above question, a novel Cross-Modality Double Bidirectional Interaction and Fusion Network (CMDBIF-Net) for RGB-T SOD is proposed. Specifically, we construct an interactive branch to indirectly bridge the RGB and thermal modalities. In addition, we propose a double bidirectional interaction (DBI) module composed of a forward interaction block (FIB) and a backward interaction block (BIB) to reduce the cross-modality differences. Moreover, a multi-scale feature enhancement and fusion (MSFEF) module is introduced to integrate the multi-modal features with considering the internal gap of different modality. Finally, we use a cascaded decoder and a cross-level feature enhancement (CLFE) module to generate high-quality saliency map. Extensive experiments are conducted on three publicly available RGB-T SOD datasets shows that the proposed CMDBIF-Net achieves outstanding performance against the state-of-the-art (SOTA) RGB-T SOD methods.

Feature fusion-based text information mining method for natural scenes

Abstract As a crucial medium of information dissemination, text holds a pivotal role in a multitude of applications. However, text detection in complex and unstructured environments presents significant challenges, such as the presence of cluttered backgrounds, variations in appearance, and uneven lighting conditions. To address this issue, this study proposes a text detection framework that leverages multistage edge detection and contextual information. This framework deviates from traditional approaches by incorporating four primary processing steps, including text visual saliency region detection to accentuate the text regions and diminish background interference, multistage edge detection to enhance the conventional stroke width transform results, a texture-based and connected components-based integration to accurately distinguish text from the background, and a context fusion step to recover missing text regions and improve the recall of text detection. The proposed method was evaluated on two widely used benchmark datasets, i.e., the international conference on document analysis and recognition (ICDAR) 2005 dataset and the ICDAR 2011 dataset, and the results indicate the advancedness of the method.

Adaptive local adversarial attacks on 3D point clouds

Modern artificial intelligence systems rely heavily on deep learning techniques. However, deep neural networks are easily disturbed by adversarial objects. Adversarial examples are beneficial to improve the robustness of the 3D neural network model and enhance the stability of the artificial intelligence system. At present, most 3D adversarial attack methods perturb the entire point cloud to generate adversarial examples, which results in high perturbation costs and low operability in the physical world. In this paper, we propose an adaptive local adversarial attack method (AL-Adv) on 3D point clouds to generate adversarial point clouds. First, we analyze the vulnerability of the 3D network model and extract the salient regions of the input point cloud, namely the vulnerable regions. Second, we propose an adaptive gradient attack algorithm that targets salient regions. The proposed attack algorithm adaptively assigns different disturbances in different directions of the three-dimensional coordinates of the point cloud. Experimental results show that our proposed AL-Adv method achieves a higher attack success rate than the global attack method. Specifically, the adversarial examples generated by AL-Adv demonstrate good imperceptibility and small generation costs.

Environmental sound classification based on improved compact bilinear attention network

Aiming at the difficulty in effectively extracting representative features from environmental sounds characterized by the complicated and considerable variation in the time-frequency domain characteristics, an improved compact bilinear attention network (ICBA-Net) is proposed to more precisely focus on the regions that need attention in feature maps in this paper. Firstly, log-mel spectrogram (LMS) extracted from environmental sound is fed into two parallel pre-trained densely connected convolutional network (DenseNet) for deep feature extraction. Afterwards, a coordinate attention module (CAM) is utilized in each branch to integrate key information in both horizontal and vertical directions to achieve a more accurate focus on global feature salient regions. Finally, two-stream enhanced global representations are integrated into a compact bilinear vector, followed by the full connection layer with activation function softmax to output the final prediction result. Experiments on two publicly available datasets (ESC-10 and ESC-50) and some environmental sounds collected from common scenarios in our real life are used to verify the validity of the proposed method. Through comparison and verification, the results show that the proposed network outperformed almost all state-of-the-art systems regarding classification accuracy, and the feasibility and effectiveness of CAM have also been verified.