Search

Based on studies using high-medium resolution images, convolutional neural networks (CNNs) and semantic segmentation have shown superiority over classical machine learning (ML), particularly in small-scale mapping. However, few/no studies have assessed the techniques on coarse resolution image classification for extensive area land cover mapping. In this study, we evaluated the performance and feasibility of three CNN models (1-D CNN, 2-D CNN, and 3-D CNN), and U-net for coarse-resolution satellite image classification and compared them to a random forest (RF) classifier. We utilized time-series, coarse resolution (1 km) composite imageries acquired by FengYun-3C visible and infrared radiometer. Labeled datasets were collected as shapefiles and split into three independent datasets: training, validation, and test datasets, and preprocessed to meet each model's input format requirements. We conducted several experiments to optimize models and select the best models. Then, the best models were evaluated on an unseen dataset. Among the DL models, one-dimensional (1-D) CNN achieved the highest overall accuracy (OA) 0. 87 and kappa (k) 0.84, 2% higher than the best results attained by 2-D CNN, 3-D CNN, and U-net models. However, 1-D CNN is outperformed by RF which achieved 0.89 (OA) and 0.87 (k). Achieving the best and the second-best results using RF and 1-D CNN models, respectively, indicates the superiority of the pixel-based method and the insignificance of spatial information in coarse-resolution image classification. Furthermore, although the DL models can yield high accuracy, especially 1-D CNN, they are less feasible than RF classifiers for coarse-resolution satellite image classification in extensive area land cover mapping.

Abstract The accurate detection of tunnel lining cracks and prompt identification of their primary causes are critical for maintaining tunnel availability. The advancement of deep learning, particularly in the domain of convolutional neural network (CNN) for image segmentation, has made tunnel lining crack detection more feasible. However, the CNN-based technique for tunnel lining crack detection commonly prioritizes increasing algorithmic complexity to enhance detection accuracy, posing a challenge in balancing the accuracy of detection and the efficiency of the algorithm. Motivated by the superior performance of Unet in image segmentation, this paper proposes a lightweight tunnel lining crack detection model named Mini-Unet, which refined the Unet architecture and utilized depthwise separable convolutions (DSConv) to replace some standard convolution layers. In the optimization of the proposed model parameters, applying a hybrid loss function that integrated dice loss and cross-entropy loss effectively tackled the imbalance between crack and background categories. Several models were set up to contrast with Mini-Unet and the experimental results were analyzed. Mini-Unet achieves a mean intersection over union (MIoU) of 60.76%, a mean precision of 84.18%, and a frame per second (FPS) of 5.635, respectively. Mini-Unet outperforms several mainstream models, enabling rapid detection while maintaining identified accuracy and facilitating the practical application of AI power for real-time tunnel lining crack detection.

The accurate detection of tunnel lining cracks and prompt identification of their primary causes are critical for maintaining tunnel availability. The advancement of deep learning, particularly in the domain of convolutional neural network (CNN) for image segmentation, has made tunnel lining crack detection more feasible. However, the CNN-based technique for tunnel lining crack detection commonly prioritizes increasing algorithmic complexity to enhance detection accuracy, posing a challenge in balancing the accuracy of detection and the efficiency of the algorithm. Motivated by the superior performance of Unet in image segmentation, this paper proposes a lightweight tunnel lining crack detection model named Mini-Unet, which refined the Unet architecture and utilized depthwise separable convolutions (DSConv) to replace some standard convolution layers. In the optimization of the proposed model parameters, applying a hybrid loss function that integrated dice loss and cross-entropy loss effectively tackled the imbalance between crack and background categories. Several models were set up to contrast with Mini-Unet and the experimental results were analyzed. Mini-Unet achieves a mean intersection over union (MIoU) of 60.76%, a mean precision of 84.18%, and a frame per second (FPS) of 5.635, respectively. Mini-Unet outperforms several mainstream models, enabling rapid detection while maintaining identified accuracy and facilitating the practical application of AI power for real-time tunnel lining crack detection. [ABSTRACT FROM AUTHOR]

A coverage prediction model helps network operators find coverage gaps, plan base station locations, evaluate quality of service, and build radio maps for spectrum sharing, interference management, localization, etc. Existing coverage prediction models rely on the height and transmission power of the base station, or the assistance of a path loss model. All of these increase the complexity of large-scale coverage predictions. In this paper, we propose a multi-modal model, DNN-SS, which combines a DNN (deep neural network) and SS (semantic segmentation) to perform coverage prediction for mobile networks. Firstly, DNN-SS filters the samples with a geospatial-temporal moving average filter algorithm, and then uses a DNN to extract numerical features. Secondly, a pre-trained model is used to perform semantic segmentation of satellite images of the measurement area. Thirdly, a DNN is used to extract features from the results after semantic segmentation to form environmental features. Finally, the prediction model is trained on the dataset consisting of numerical features and environmental features. The experimental results on campus show that for random location prediction, the model achieves a RMSE (Root Mean Square Error) of 1.97 dB and a MAE (Mean Absolute Error) of 1.41 dB, which is an improvement of 10.86% and 10.2%, respectively, compared with existing models. For the prediction of a test area, the RMSE and MAE of the model are 4.32 dB and 3.45 dB, respectively, and the RMSE is only 0.22 dB lower than that of existing models. However, the DNN-SS model does not need the height, transmission power, and antenna gain of the base station, or a path loss model, which makes it more suitable for large-scale coverage prediction. [ABSTRACT FROM AUTHOR]

To address the cost issue associated with pixel-level image annotation in fully supervised semantic segmentation, a method based on semi-supervised semantic segmentation is proposed for extracting winter wheat planting areas. This approach utilizes self-training with pseudo-labels to learn from a small set of images with pixel-level annotations and a large set of unlabeled images, thereby achieving the extraction. In the constructed initial dataset, a random sampling strategy is employed to select 1/16, 1/8, 1/4, and 1/2 proportions of labeled data. Furthermore, in conjunction with the concept of consistency regularization, strong data augmentation techniques are applied to the unlabeled images, surpassing classical methods such as cropping and rotation to construct a semi-supervised model. This effectively alleviates overfitting caused by noisy labels. By comparing the prediction results of different proportions of labeled data using SegNet, DeepLabv3+, and U-Net, it is determined that the U-Net network model yields the best extraction performance. Moreover, the evaluation metrics MPA and MIoU demonstrate varying degrees of improvement for semi-supervised semantic segmentation compared to fully supervised semantic segmentation. Notably, the U-Net model trained with 1/16 labeled data outperforms the models trained with 1/8, 1/4, and 1/2 labeled data, achieving MPA and MIoU scores of 81.63%, 73.31%, 82.50%, and 76.01%, respectively. This method provides valuable insights for extracting winter wheat planting areas in scenarios with limited labeled data. [ABSTRACT FROM AUTHOR]

Rice lodging seriously affects rice quality and production. Traditional manual methods of detecting rice lodging are labour-intensive and can result in delayed action, leading to production loss. With the development of the Internet of Things (IoT), unmanned aerial vehicles (UAVs) provide imminent assistance for crop stress monitoring. In this paper, we proposed a novel lightweight detection system with UAVs for rice lodging. We leverage UAVs to acquire the distribution of rice growth, and then our proposed global attention network (GloAN) utilizes the acquisition to detect the lodging areas efficiently and accurately. Our methods aim to accelerate the processing of diagnosis and reduce production loss caused by lodging. The experimental results show that our GloAN can lead to a significant increase in accuracy with negligible computational costs. We further tested the generalization ability of our GloAN and the results show that the GloAN generalizes well in peers' models (Xception, VGG, ResNet, and MobileNetV2) with knowledge distillation and obtains the optimal mean intersection over union (mIoU) of 92.85%. The experimental results show the flexibility of GloAN in rice lodging detection. [ABSTRACT FROM AUTHOR]

As an ecosystem in transition from land to sea, mangroves play a vital role in wind and wave protection and biodiversity maintenance. However, the invasion of Spartina alterniflora Loisel seriously damages the mangrove wetland ecosystem. To protect mangroves scientifically and dynamically, a semantic segmentation model for mangroves and Spartina alterniflora Loise was proposed based on UperNet (Swin-UperNet). In the proposed Swin-UperNet model, a data concatenation module was proposed to make full use of the multispectral information of remote sensing images, the backbone network was replaced with a Swin transformer to improve the feature extraction capability, and a boundary optimization module was designed to optimize the rough segmentation results. Additionally, a linear combination of cross-entropy loss and Lovasz-Softmax loss was taken as the loss function of Swin-UperNet, which could address the problem of unbalanced sample distribution. Taking GF-1 and GF-6 images as the experiment data, the performance of the Swin-UperNet model was compared against that of other segmentation models in terms of pixel accuracy (PA), mean intersection over union (mIoU), and frames per second (FPS), including PSPNet, PSANet, DeepLabv3, DANet, FCN, OCRNet, and DeepLabv3+. The results showed that the Swin-UperNet model achieved the best PA of 98.87% and mIoU of 90.0%, and the efficiency of the Swin-UperNet model was higher than that of most models. In conclusion, Swin-UperNet is an efficient and accurate model for mangrove and Spartina alterniflora Loise segmentation synchronously, which will provide a scientific basis for Spartina alterniflora Loise monitoring and mangrove resource conservation and management. [ABSTRACT FROM AUTHOR]

Early production warnings are usually labor-intensive, even with remote sensing techniques in highly intensive but fragmented growing areas with various phenological stages. This study used high-resolution unmanned aerial vehicle (UAV) images with a ground sampling distance (GSD) of 3 cm to detect the plant body of pineapples. The detection targets were mature fruits mainly covered with two kinds of sun protection materials—round plastic covers and nets—which could be used to predict the yield in the next two to three months. For round plastic covers (hereafter referred to as wearing a hat), the Faster R-CNN was used to locate and count the number of mature fruits based on input image tiles with a size of 256 × 256 pixels. In the case of intersection-over-union (IoU) > 0.5, the F1-score of the hat wearer detection results was 0.849, the average precision (AP) was 0.739, the precision was 0.990, and the recall was 0.743. We used the Mask R-CNN model for other mature fruits to delineate the fields covered with nets based on input image tiles with a size of 2000 × 2000 pixels and a mean IoU (mIoU) of 0.613. Zonal statistics summed up the area with the number of fields wearing a hat and covered with nets. Then, the thresholding procedure was used to solve the potential issue of farmers' harvesting in different batches. In pineapple cultivation fields, the zonal results revealed that the overall classification accuracy is 97.46%, and the kappa coefficient is 0.908. The results were expected to demonstrate the critical factors of yield estimation and provide researchers and agricultural administration with similar applications to give early warnings regarding production and adjustments to marketing. [ABSTRACT FROM AUTHOR]

Hyperspectral image (HSI) classification is one of the most important remote sensing techniques. Currently, the performances of most of the HSI classification networks on the public HSI datasets are overoptimistic (i.e., the overall accuracy exceeds 98 %). This deficiency is partly due to the very limited scale of these existing datasets, which impedes the network generalization and poses limitations for the future development. The existing hyperspectral datasets urgently need to be scaled up in size. Therefore, in this study, we built a dataset named the WHU-OHS dataset, which consists of about 90 million manually labeled samples of 7795 Orbita hyperspectral satellite (OHS) image patches (sized 512 × 512) from 40 Chinese locations. This dataset ranges from the visible to near-infrared range, with an average spectral resolution of 15 nm. The extensive geographical distribution, large spatial coverage, and widely used classification system make the WHU-OHS dataset a challenging benchmark. This dataset was validated by comprehensive experiments using several representative deep HSI classification networks. Furthermore, the transferability of the HSI classification networks under the conditions of the same/different HSI sensors was tested. In particular, when classifying the existing public HSI datasets, using initial parameters obtained by pre-training on the WHU-OHS dataset can further improve the inference accuracy as well as the training efficiency. The WHU-OHS dataset and a PyTorch toolbox for large-scale HSI classification are available at https://irsip.whu.edu.cn/resources/resources_v2.php and https://github.com/zjjerica/WHU-OHS-Pytorch, respectively.

Copyright of Tourism Tribune / Lvyou Xuekan is the property of Tourism Institute of Beijing Union University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

At present, road continuity is a major challenge, and it is difficult to extract the centerline vector of roads, especially when the road view is obstructed by trees or other structures. Most of the existing research has focused on optimizing the available deep-learning networks. However, the segmentation accuracy is also affected by the loss function. Currently, little research has been published on road segmentation loss functions. To resolve this problem, an attention loss function named GapLoss that can be combined with any segmentation network was proposed. Firstly, a deep-learning network was used to obtain a binary prediction mask. Secondly, a vector skeleton was extracted from the prediction mask. Thirdly, for each pixel, eight neighboring pixels with the same value of the pixel were calculated. If the value was 1, then the pixel was identified as the endpoint. Fourth, according to the number of endpoints within a buffered range, each pixel in the prediction image was given a corresponding weight. Finally, the weighted average value of the cross-entropy of all the pixels in the batch was used as the final loss function value. We employed four well-known semantic segmentation networks to conduct comparative experiments on three large datasets. The results showed that, compared to other loss functions, the evaluation metrics after using GapLoss were nearly all improved. From the predicted image, the road prediction by GapLoss was more continuous, especially at intersections and when the road was obscured from view, and the road segmentation accuracy was improved. [ABSTRACT FROM AUTHOR]

Rice lodging seriously affects rice quality and production. Traditional manual methods of detecting rice lodging are labour-intensive and can result in delayed action, leading to production loss. With the development of the Internet of Things (IoT), unmanned aerial vehicles (UAVs) provide imminent assistance for crop stress monitoring. In this paper, we proposed a novel lightweight detection system with UAVs for rice lodging. We leverage UAVs to acquire the distribution of rice growth, and then our proposed global attention network (GloAN) utilizes the acquisition to detect the lodging areas efficiently and accurately. Our methods aim to accelerate the processing of diagnosis and reduce production loss caused by lodging. The experimental results show that our GloAN can lead to a significant increase in accuracy with negligible computational costs. We further tested the generalization ability of our GloAN and the results show that the GloAN generalizes well in peers’ models (Xception, VGG, ResNet, and MobileNetV2) with knowledge distillation and obtains the optimal mean intersection over union (mIoU) of 92.85%. The experimental results show the flexibility of GloAN in rice lodging detection.

The capability to understand visual scenes with limited labeled data has been widely concerned in the field of computer vision. Although semi-supervised learning for image classification has been extensively studied in some cases, semantic segmentation with limited data has only recently gained attention. In this work, we follow the standard semi-supervised segmentation pipeline for image classification and propose a two-branch network that can encode strong and pseudo label spaces respectively, extracting reliable supervision information from pseudo-labels to assist in training network with strong labels. Our method outperforms previous semi-supervised methods with limited annotation cost. On standard benchmark PASCAL VOC 2012 for semi-supervised semantic segmentation, the proposed approach gains fresh state-of-the-art performance.

Copyright of Journal of Data Acquisition & Processing / Shu Ju Cai Ji Yu Chu Li is the property of Editorial Department of Journal of Nanjing University of Aeronautics & Astronautics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

With the availability of low-cost depth-visual sensing devices, such as Microsoft Kinect, we are experiencing a growing interest in indoor environment understanding, at the core of which is semantic segmentation in RGB-D image. The latest research shows that the convolutional neural network (CNN) still dominates the image semantic segmentation field. However, down-sampling operated during the training process of CNNs leads to unclear segmentation boundaries and poor classification accuracy. To address this problem, in this paper, we propose a novel end-to-end deep architecture, termed FuseCRFNet, which seamlessly incorporates a fully-connected Conditional Random Fields (CRFs) model into a depth-based CNN framework. The proposed segmentation method uses the properties of pixel-to-pixel relationships to increase the accuracy of image semantic segmentation. More importantly, we formulate the CRF as one of the layers in FuseCRFNet to refine the coarse segmentation in the forward propagation, in meanwhile, it passes back the errors to facilitate the training. The performance of our FuseCRFNet is evaluated by experimenting with SUN RGB-D dataset, and the results show that the proposed algorithm is superior to existing semantic segmentation algorithms with an improvement in accuracy of at least 2%, further verifying the effectiveness of the algorithm. [ABSTRACT FROM AUTHOR]

Copyright of Chinese Medical Equipment Journal is the property of Chinese Medical Equipment Journal Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In this paper, a novel semi-supervised segmentation framework based on a spot-divergence supervoxelization of multi-sensor fusion data is proposed for autonomous forest machine (AFMs) applications in complex environments. Given the multi-sensor measuring system, our framework addresses three successive steps: firstly, the relationship of multi-sensor coordinates is jointly calibrated to form higher-dimensional fusion data. Then, spot-divergence supervoxels representing the size-change property are given to produce feature vectors covering comprehensive information of multi-sensors at a time. Finally, the Gaussian density peak clustering is proposed to segment supervoxels into sematic objects in the semi-supervised way, which non-requires parameters preset in manual. It is demonstrated that the proposed framework achieves a balancing act both for supervoxel generation and sematic segmentation. Comparative experiments show that the well performance of segmenting various objects in terms of segmentation accuracy (F-score up to 95.6%) and operation time, which would improve intelligent capability of AFMs. [ABSTRACT FROM AUTHOR]

• The recurrent residual convolutional neural network can extract more features of fine cracks. • The context extractor network can retain more spatial information of the crack image and capture more crack features. • R2CE-Net model can eliminate noise interference and extract clearer crack edges. • The software interface can display various crack detection information intuitively. To cope with the complex bridge crack detection environment, we developed a flexible crack identification system. Firstly, the acquired images are processed by sliding window technology to construct a bridge crack dataset. Then we propose a trainable context encoder network which uses the recurrent residual convolutional neural network (RRCNN) to improve the encoder structure to better extract low-level features. Additionally, it combines dense atrous convolution block (DAC) and residual multi-kernel pooling block (RMP) which can retain more crack information and features from the crack image, as well as improve the performance of crack segmentation. Finally, our research results are validated in the developed software platform. The experiments show that our method is more stable and accurate than other methods, its accuracy and mean intersection over union (mIoU) have reached 98.62% and 80.93%, respectively. Moreover, it performs well in cracks facing complex conditions, which can be applied to the bridge maintenance projects. [ABSTRACT FROM AUTHOR]

• Sample-wise classification of ECG waves from continuous raw noisy signals. • Channel-mix convolutional to handle lead-to lead variabilities. • Hybrid model for automatic extraction and classification of samples by considering forward and backward time stamps. • Experiments with the noisy dataset to increase the robustness, reliability and generalization. • There are 5.77-12.28 % increments in weighted average accuracy using the proposed hybrid model than other LSTM networks. Interpretation of the ECG waves plays a vital role in analysis of cardiovascular diseases. Therefore, many semi and fully-automatic approaches using advanced machine learning techniques for the ECG waves detection are being exhaustively investigated by the researchers. Regardless of advanced machine learning or deep learning techniques, present methods lacks generalization, robustness, reliability and real-time implementation. Moreover, none of the existing study had presented the information related to lead variability which is imperative aspects to handle for accurate detection of ECG waves. Hence, in this paper, we are presenting the new approach of ECG wave segmentation called semantic segmentation, a well-known concept in image segmentation. The proposed approach mainly includes; 1) hybrid channel-mix convolutional and bidirectional LSTM which able to extract temporal dependencies as well as short and long-time dependencies in forward and backward time stamps for semantic segmentation. 2) Experiments with channel-mix convolution to handle lead to lead variability, 3) Experiments with noisy dataset to increase the robustness, reliability and generalization. The proposed hybrid model is implemented on the standard publicly available QT database. The remarkable results with high accuracy ranging from 96 to 98.56 % with average and weighted accuracies of 96.72 % and 95.54 % respectively are obtained for segmentation of ECG waves from continuous raw ECG signal. There are 12.28 %, 8.08 % and 5.77 % increments in weighted average accuracy using proposed hybrid model than LSTM, BiLSTM and double BiLSTM respectively. [ABSTRACT FROM AUTHOR]