A2SPPNet: Attentive Atrous Spatial Pyramid Pooling Network for Salient Object Detection

While deep learning-based methods have gained considerable improvements in remote sensing (RS) image change detection (CD), scale variations and pseudochanges hinder most supervised methods’ performance. The CD networks derived from other fields can be fronted with false alarms and miss detections in high-resolution RS images due to the weak feature representation ability. In this article, an attention-guided end-to-end change detection network (AGCDetNet) is proposed based on the fully convolutional network and attention mechanism. AGCDetNet learns to enhance the feature representation of change information and achieve accuracy improvements using spatial attention and channel attention. A spatial attention module (SPAM) promotes the discrimination between the changed objects and the background by adding the learned spatial attention to the deep features. Channelwise attention-guided interference filtering unit (CIFU)/atrous spatial pyramid pooling (CG-ASPP) module enhances the representation of multilevel features and multiscale context, respectively. Extensive experiments have been conducted on several public datasets for performance evaluation, including LEVIR-CD, WHU, Season-Varying, WV2, and ZY3. Experiment results demonstrate that AGCDetNet outperforms several state-of-the-art methods of accuracy and robustness. Specifically, AGCDetNet achieves the best F1-score on two datasets, i.e., LEVIR-CD (0.9076) and Season-Varying (0.9654).

This study aims to enhance the efficiency and accuracy of thyroid nodule segmentation in ultrasound images, ultimately improving nodule detection and diagnosis. For clinical deployment on mobile and embedded devices, DeepLabV3+ strives to achieve a balance between a lightweight architecture and high segmentation accuracy. A comprehensive dataset of ultrasound images was meticulously curated using a high-resolution ultrasound imaging device. Data acquisition adhered to standardized protocols to ensure high-quality imaging. Preprocessing steps, including noise reduction and contrast optimization, were applied to enhance image clarity. Expert radiologists provided ground truth labels through meticulous annotation. To improve segmentation performance, we integrated MobileNetV2 and Depthwise Separable Dilated Convolution into the Atrous Spatial Pyramid Pooling (ASPP) module, incorporating the Pyramid Pooling Module (PPM) and attention mechanisms. To mitigate classification imbalances, we employed Tversky loss functions in the ultrasound image classification process. In semantic image segmentation, DeepLabV3+ achieved an impressive Intersection over Union (IoU) of 94.37%, while utilizing only 12.4 MB of parameters, including weights and biases. This remarkable accuracy demonstrates the effectiveness of our approach. A high IoU value in medical imaging analysis reflects the model's ability to accurately delineate object boundaries. DeepLabV3+ represents a significant advancement in thyroid nodule segmentation, particularly for thyroid cancer screening and diagnosis. The obtained segmentation results suggest promising directions for future research, especially in the early detection of thyroid nodules. Deploying this algorithm on mobile devices offers a practical solution for early diagnosis and is likely to improve patient outcomes.

In light of recent advances in deep learning and Synthetic Aperture Radar (SAR) technology, there has been a growing adoption of ship detection models that are based on deep learning methodologies. However, the efficiency of SAR ship detection models is significantly impacted by complex backgrounds, noise, and multi-scale ships (the number of pixels occupied by ships in SAR images varies significantly). To address the aforementioned issues, this research proposes a Pyramid Pooling Attention Network (PPA-Net) for SAR multi-scale ship detection. Firstly, a Pyramid Pooled Attention Module (PPAM) is designed to alleviate the influence of background noise on ship detection while its parallel component favors the processing of multiple ship sizes. Different from the previous attention module, the PPAM module can better suppress the background noise in SAR images because it considers the saliency of ships in SAR images. Secondly, an Adaptive Feature Balancing Module (AFBM) is developed, which can automatically balance the conflict between ship semantic information and location information. Finally, the detection capabilities of the ship detection model for multi-scale ships are further improved by introducing the Atrous Spatial Pyramid Pooling (ASPP) module. This innovative module enhances the detection model’s ability to detect ships of varying scales by extracting features from multiple scales using atrous convolutions and spatial pyramid pooling. PPA-Net achieved detection accuracies of 95.19% and 89.27% on the High-Resolution SAR Images Dataset (HRSID) and the SAR Ship Detection Dataset (SSDD), respectively. The experimental results demonstrate that PPA-Net outperforms other ship detection models.

A pooling-based feature pyramid network for salient object detection

Salient object detection (SOD) has achieved remarkable performance in well‐illuminated scenarios. However, in inevitable low‐light environments, SOD faces inherent challenges such as insufficient illumination, noise interference, and contrast degradation, which severely compromise the perceptual accuracy of visual models. To address these issues, we propose a novel Edge‐Guided Dual‐Stream Collaborative Network (EGDSNet), which enhances the integrity of salient objects through multi‐level feature interaction between edge detection and segmentation branches. Specifically, EGDSNet adopts a dual‐branch architecture: the edge detection branch integrates a Local Context Refinement (LCR) module and a Cross‐Modal Edge Refinement (CER) module to suppress noise while reinforcing boundary responses of targets. The segmentation branch employs a Global Feature Fusion (GFF) module and an Edge‐Guided Feature Enhancement (EFE) module to achieve complementary integration of global semantics and local details. The two branches interact via an edge‐guided attention mechanism, enabling saliency predictions to maintain structural integrity while attaining precise boundary localization. Additionally, the Residual Channel Attention (RCA) module is leveraged to further suppress noise and extract structural details from low‐level features. Comprehensive experiments on the LLI and SOD‐LL datasets demonstrate that our model outperforms state‐of‐the‐art methods in both quantitative metrics and qualitative visualizations, validating its superiority in low‐light SOD tasks.

The purpose of human object detection is to obtain the number of people and their position in images, which is one of the core problems in the field of machine vision. However, the high missing detection rate from small- and medium-sized human bodies due to the large variety of human scale in human object detection tasks still influences the performance of human object detection. To solve the above problem, this paper proposed an improved ASPP_BiFPN_YOLOv4 (ABYOLOv4) method to detect human object detection. In detail, Atrous Spatial Pyramid Pooling (ASPP) module was used to replace the original Spatial Pyramid Pooling module to increase the receptive field level of the network and improve the perception ability of multi-scale targets. Then, the original Path Aggregation Network (PANet) multi-scale fusion module was replaced by the self-built bi-layer bidirectional feature pyramid network (Bi-FPN). Meanwhile, a new feature was imported into the proposed model to reuse the mid- and low-level features, which could enhance the ability of the network to express the characteristics of small- and medium-sized targets. Finally, the standard convolution in Bi-FPN was replaced by depth-separable convolution to make the network achieve the balance of accuracy and the number of parameters. To identify the performance of the proposed ABYOLOv4 model, the human object detection experiment is carried out by using the public data set of VOC2007 and VOC2012, the improved YOLOv4 algorithm is 0.5% higher than the original AP algorithm, and the weight file size of the model is reduced by 45.3 M. The experimental results demonstrated that the proposed ABYOLOv4 network has higher accuracy and lower computational cost for human target detection.

Attention and boundary guided salient object detection

Natural breeding scenes have the characteristics of a large number of cows, complex lighting, and a complex background environment, which presents great difficulties for the detection of dairy cow estrus behavior. However, the existing research on cow estrus behavior detection works well in ideal environments with a small number of cows and has a low inference speed and accuracy in natural scenes. To improve the inference speed and accuracy of cow estrus behavior in natural scenes, this paper proposes a cow estrus behavior detection method based on the improved YOLOv5. By improving the YOLOv5 model, it has stronger detection ability for complex environments and multi-scale objects. First, the atrous spatial pyramid pooling (ASPP) module is employed to optimize the YOLOv5l network at multiple scales, which improves the model’s receptive field and ability to perceive global contextual multiscale information. Second, a cow estrus behavior detection model is constructed by combining the channel-attention mechanism and a deep-asymmetric-bottleneck module. Last, K-means clustering is performed to obtain new anchors and complete intersection over union (CIoU) is used to introduce the relative ratio between the predicted box of the cow mounting and the true box of the cow mounting to the regression box prediction function to improve the scale invariance of the model. Multiple cameras were installed in a natural breeding scene containing 200 cows to capture videos of cows mounting. A total of 2668 images were obtained from 115 videos of cow mounting events from the training set, and 675 images were obtained from 29 videos of cow mounting events from the test set. The training set is augmented by the mosaic method to increase the diversity of the dataset. The experimental results show that the average accuracy of the improved model was 94.3%, that the precision was 97.0%, and that the recall was 89.5%, which were higher than those of mainstream models such as YOLOv5, YOLOv3, and Faster R-CNN. The results of the ablation experiments show that ASPP, new anchors, C3SAB, and C3DAB designed in this study can improve the accuracy of the model by 5.9%. Furthermore, when the ASPP dilated convolution was set to (1,5,9,13) and the loss function was set to CIoU, the model had the highest accuracy. The class activation map function was utilized to visualize the model’s feature extraction results and to explain the model’s region of interest for cow images in natural scenes, which demonstrates the effectiveness of the model. Therefore, the model proposed in this study can improve the accuracy of the model for detecting cow estrus events. Additionally, the model’s inference speed was 71 frames per second (fps), which meets the requirements of fast and accurate detection of cow estrus events in natural scenes and all-weather conditions.

Synthetic Aperture Radar (SAR) can provide rich feature information under all-weather and day-night conditions because it is not affected by climatic conditions. However, multiplicative speckle noise exists in SAR images, which makes it difficult to accurately identify some fuzzy targets in SAR images, such as roads and rivers, during semantic segmentation. This paper proposes an improved Deeplabv3+ network that can be effectively applied to the semantic segmentation task of SAR images. Firstly, this paper added the attention mechanism and, combined with the idea of an image pyramid, proposed the Feature Post-Processing Module (FPPM) to post-process the network output feature map, obtain better fine image features, and solve the problem of fuzzy texture and spectral features of SAR images. Compared to the original Deeplabv3+ network, the segmentation accuracy has been improved by 3.64% and mIoU improved by 1.09%. Secondly, to solve the problems of limited SAR image data and an unbalanced sample, this paper used the focal loss function to improve the backbone function of the network, which increased the mIoU by 1.01%. Finally, the Atrous Spatial Pyramid Pooling (ASPP) module was improved and the 3 × 3 void convolution in ASPP was decomposed into 2D, which can maintain the void ratio and effectively reduce the calculation amount of the module, shorten the training time by 19 ms and improve the semantic segmentation effect.

Given that existing salient object detection methods cannot effectively predict the fine contours of salient objects when extracting local or global contexts and features, we propose a novel contour self-compensated network (CSCNet) to generate a more accurate saliency map with complete contour. Unlike the common binary saliency detection, we reconstruct the salient object detection problem into a multi-classification problem of the background, the salient object, and the salient contour, where the salient contour is used as the third label for ground truth. Meanwhile, the image and its superpixel map are concatenated as the input of our network to add more edge information. Also, a penalty loss is proposed to restrict the spatial relationship between the background, objects, and their contours. Experimentally, we evaluate the proposed CSCNet on six benchmark datasets in both accuracy and efficiency and evaluate the attribute-based performance on the SOC dataset. Compared with 13 state-of-the-art algorithms, our CSCNet can detect salient objects more accurately and completely without adding too many convolutional layers and parameters.

Traditional DeepLabV3+ image semantic segmentation methods face challenges in pitaya orchard environments characterized by multiple interference factors, complex image backgrounds, high computational complexity, and extensive memory consumption. This paper introduces an improved visual navigation path recognition method for pitaya orchards. Initially, DeepLabV3+ utilizes a lightweight MobileNetV2 as its primary feature extraction backbone, which is augmented with a Pyramid Split Attention (PSA) module placed after the Atrous Spatial Pyramid Pooling (ASPP) module. This improvement enhances the spatial feature representation of feature maps, thereby sharpening the segmentation boundaries. Additionally, an Efficient Channel Attention Network (ECANet) mechanism is integrated with the lower-level features of MobileNetV2 to reduce computational complexity and refine the clarity of target boundaries. The paper also designs a navigation path extraction algorithm, which fits the road mask regions segmented by the model to achieve precise navigation path recognition. Experimental findings show that the enhanced DeepLabV3+ model achieved a Mean Intersection over Union (MIoU) and average pixel accuracy of 95.79% and 97.81%, respectively. These figures represent increases of 0.59 and 0.41 percentage points when contrasted with the original model. Furthermore, the model’s memory consumption is reduced by 85.64%, 84.70%, and 85.06% when contrasted with the Pyramid Scene Parsing Network (PSPNet), U-Net, and Fully Convolutional Network (FCN) models, respectively. This reduction makes the proposed model more efficient while maintaining high segmentation accuracy, thus supporting enhanced operational efficiency in practical applications. The test results for navigation path recognition accuracy reveal that the angle error between the navigation centerline extracted using the least squares method and the manually fitted centerline is less than 5°. Additionally, the average deviation between the road centerlines extracted under three different lighting conditions and the actual road centerline is only 2.66 pixels, with an average image recognition time of 0.10 s. This performance suggests that the study can provide an effective reference for visual navigation in smart agriculture.

Many existing salient object detection methods are dedicated to fusing features from different levels of a pre-trained convolutional neural network (CNN). However, these methods can easily lead to internal discontinuities within the salient objects because of unreasonable feature fusion strategies and short-range dependencies resulting from common convolution and pooling operations. In this paper, we propose a novel non-local duplicate pooling (NLDP) network to overcome these internal discontinuities. NLDP begins by removing the first few convolutional layers of a classic CNN, which have small receptive fields and require large amounts of calculation. A novel duplicate pooling module (DPM) is then used to generate richer and more detailed saliency maps. This is achieved by constructing a double-pathway that can integrating partial feature maps. Within the DPM, a non-local module (NLM) is used to obtain long-range dependencies. This enhances the internal continuities between the saliency maps. Comprehensive experiments conducted on six benchmark datasets have confirmed the increased effectiveness and detection speed of our method in relation to other salient object detection methods.

DSA-Net: A novel deeply supervised attention-guided network for building change detection in high-resolution remote sensing images

Salient object detection (SOD) aims to determine the most visually attractive objects in an image. With the development of virtual reality (VR) technology, 360° omnidirectional image has been widely used, but the SOD task in 360° omnidirectional image is seldom studied due to its severe distortions and complex scenes. In this article, we propose a multi-projection fusion and refinement network (MPFR-Net) to detect the salient objects in 360° omnidirectional image. Different from the existing methods, the equirectangular projection (EP) image and four corresponding cube-unfolding (CU) images are embedded into the network simultaneously as inputs, where the CU images not only provide supplementary information for EP image but also ensure the object integrity of cube-map projection. In order to make full use of these two projection modes, a dynamic weighting fusion (DWF) module is designed to adaptively integrate the features of different projections in a complementary and dynamic manner from the perspective of inter and intrafeatures. Furthermore, in order to fully explore the way of interaction between encoder and decoder features, a filtration and refinement (FR) module is designed to suppress the redundant information of the feature itself and between the features. Experimental results on two omnidirectional datasets demonstrate that the proposed approach outperforms the state-of-the-art methods both qualitatively and quantitatively. The code and results can be found from the link of https://rmcong.github.io/proj_MPFRNet.html.

As the key structure connecting the vegetative and reproductive organs of soybean plants, the main stem plays a crucial role, and its morphological parameters serve as core phenotypic indicators for evaluating plant growth, lodging resistance, and yield potential. At the mature stage, the main stem exhibits high similarity to pods in color and texture, along with complex curvature and severe occlusion by pods and leaves, making accurate and continuous extraction challenging for conventional segmentation methods. To address this, this study proposes RAM-UNet, a high-precision semantic segmentation model based on an improved U-Net architecture. The model adopts ResNet50 as the backbone and replaces standard convolutions with deformable convolutions to capture curved stem morphology and improve feature extraction for low-contrast edges. In the encoder, the Convolutional Block Attention Module (CBAM) is combined with an improved atrous spatial pyramid pooling (ASPP) module (C-ASPP) with four dilation rates, enhancing multi-scale feature representation compared to the original three-rate design. A multi-scale attention aggregation (MSAA) module in the decoder improves continuity and integrity of stem boundaries. During training, a composite loss function combining Dice loss and cross-entropy loss is employed to mitigate foreground pixel sparsity. Experimental results on a self-constructed dataset show that RAM-UNet achieves a mean Intersection over Union (mIoU) of 90.58%, with Recall and Precision reaching 94.99% and 94.58%, respectively. Compared with U-Net, DeepLabv3+, PSPNet, and SegNet, RAM-UNet improves mIoU by 6.41%, 10.51%, 22.41%, and 17.37%, respectively. Automatically measured stem lengths show high agreement with manual measurements (R² = 0.9746), validating practical applicability. RAM-UNet also generalizes well on the public PASCAL VOC 2012 dataset, achieving an mIoU of 73.14%. The results indicate that the proposed model enables high-precision and continuous segmentation of main stems in mature soybean plants, providing an effective technical solution for automated and non-destructive measurement of crop phenotypic parameters.