Early detection and treatment of diabetic retinopathy can delay blindness and improve quality of life for diabetic patients. It is difficult to detect early symptoms of diabetic retinopathy, which is presented by few ...
详细信息
ISBN:
(纸本)9798350396386
Early detection and treatment of diabetic retinopathy can delay blindness and improve quality of life for diabetic patients. It is difficult to detect early symptoms of diabetic retinopathy, which is presented by few microaneurysms in fundus images. This study proposes an algorithm to detect microaneurysms in fundus images automatically. The proposal includes microaneurysms segmentation by U-Net model and their false positives removal by ResNet model. The effectiveness of the proposal is evaluated with the public database IDRiD and E-ophtha by the Area Under Precision Recall curve (AUPR). 90% of microaneurysms can be detected at early stages of diabetic retinopathy. This proposal outperforms previous methods based in AUPR evaluation.
Earth observation satellites in low earth orbit (LEO) collect a large amount of image data daily, while space-to-ground links have become the major bottleneck for data transmission due to the limited bandwidth. Existi...
详细信息
ISBN:
(数字)9798331517786
ISBN:
(纸本)9798331517793
Earth observation satellites in low earth orbit (LEO) collect a large amount of image data daily, while space-to-ground links have become the major bottleneck for data transmission due to the limited bandwidth. Existing approaches focus on exploring more efficient routing strategies to achieve better data transmission but still struggle to keep pace with the surging volume of observed data. However, the advancement of onboard computing power has opened the possibility of processing data on satellites to reduce the transmitted data volume. This paper proposes a distributed deep reinforcement learning (DRL) algorithm to improve transmission efficiency by jointly optimizing computing and routing. Aiming to minimize task latency while considering the limitations of satellite storage resources, the problem is modeled as a partially observable Markov process (POMDP). An algorithm based on dueling double deep Q-Network (Dueling-DDQN) is proposed to achieve dynamic decision-making utilizing local and neighboring resource states. Furthermore, a method for dynamic optimization of backhaul destinations is proposed, using the pre-trained Q-network to estimate action values across multiple candidate destination satellites, thus enabling further optimization of data transmission without additional training. Simulation results indicate that the proposed algorithm achieves the lowest latency across various task loads compared to baseline methods.
Considering large capacitance current of the high proportion cable distribution network, coupled with the influence of wind turbine access on the selection of the distribution network grounding, single-phase is only g...
详细信息
Integrated circuits based on digital imageprocessing have been increasingly important in recent years due to their usefulness in many different fields, including manufacturing, agriculture, remote sensing, and medici...
Integrated circuits based on digital imageprocessing have been increasingly important in recent years due to their usefulness in many different fields, including manufacturing, agriculture, remote sensing, and medicine. The development of low-area, low-power, and low-delay processors is a significant challenge. The majority of these efforts have targeted efficient noise removal filtering by minimizing the resources required to accomplish it. Existing filtering approaches require a lot of space, a lot of power, and a lot of time since noise is dynamic and distributed randomly throughout images. Finally, this research work is extended by incorporating QCA technology to reduce the space, power, and time needed to build an efficient filtering method with altered noise deviation features. This research study proposes a majority logic gates technology to create DCNN-based FBF on a field-programmable gate array (FPGA). By using the vast majority of logic gates to design the MCSA, considerable gains can be achieved in terms of speed, execution time, area, power, and latency over the RCA used in traditional CNN. The proposed DCNN-FBF is also used in the implementation of an image denoising application. The proposed methodology has been shown to improve hardware performance relative to state-of-the-art methodologies, as evidenced by both hardware and extensive simulation findings. In addition, as compared to traditional filtering methods, the proposed image denoising solution has achieved higher values of structural similarity (SSIM) and PSNR.
In this paper, we present KeyMatchNet, a novel network for zero-shot pose estimation in 3D point clouds. Our method uses only depth information, making it more applicable for many industrial use cases, as color inform...
详细信息
ISBN:
(数字)9798350358513
ISBN:
(纸本)9798350358520
In this paper, we present KeyMatchNet, a novel network for zero-shot pose estimation in 3D point clouds. Our method uses only depth information, making it more applicable for many industrial use cases, as color information is seldom available. The network is composed of two parallel components for computing object and scene features. The features are then combined to create matches used for pose estimation. The parallel structure allows for pre-processing of the individual parts, which decreases the *** a zero-shot network allows for a very short set-up time, as it is not necessary to train models for new objects. However, as the network is not trained for the specific object, zero-shot pose estimation methods generally have lower accuracy compared with conventional methods. To address this, we reduce the complexity of the task by including the scenario information during training. This is typically not feasible as collecting real data for new tasks drastically increases the cost. However, for zero-shot pose estimation, training for new objects is not necessary and the expensive data collection can thus be performed only *** method is trained on 1,500 objects and is only tested on unseen objects. We demonstrate that the trained network can not only accurately estimate poses for novel objects, but also demonstrate the ability of the network on objects outside of the trained class. Test results are also shown on real data. We believe that the presented method is valuable for many real-world scenarios. Project page available at ***
In recent years, the integration of advanced imaging techniques and deep learning methods has significantly advanced computer-aided diagnosis (CAD) systems for breast cancer detection and classification. Transformers,...
详细信息
ISBN:
(数字)9798331529710
ISBN:
(纸本)9798331529727
In recent years, the integration of advanced imaging techniques and deep learning methods has significantly advanced computer-aided diagnosis (CAD) systems for breast cancer detection and classification. Transformers, which have shown great promise in computer vision, are now being applied to medical image analysis. However, their application to histopathological images presents challenges due to the need for extensive manual annotations of whole-slide images (WSIs), as these models require large amounts of data to work effectively, which is costly and time-consuming. Furthermore, the quadratic computational cost of Vision Transformers (ViTs) is particularly prohibitive for large, high-resolution histopathological images, especially on edge devices with limited computational resources. In this study, we introduce a novel lightweight breast cancer classification approach using transformers that operates effectively without large datasets. By incorporating parallelprocessing pathways for Discrete Cosine Transform (DCT) Attention and MobileConv, we convert image data from the spatial domain to the frequency domain to utilize the benefits such as filtering out high frequencies in the image, which reduces computational cost. This demonstrates the potential of our approach to improve breast cancer classification in histopathological images, offering a more efficient solution with reduced reliance on extensive annotated datasets. Our proposed model achieves an accuracy of $\mathbf{9 6 0 0 \%} \pm \mathbf{0 4 8 \%}$ for binary classification and $\mathbf{8 7 8 5 \%} \pm 0.93 \%$ for multiclass classification, which is comparable to state-of-the-art models while significantly reducing computational costs. This demonstrates the potential of our approach to improve breast cancer classification in histopathological images, offering a more efficient solution with reduced reliance on extensive annotated datasets.
Inflammatory Bowel Disease (IBD) is a global chronic intestinal inflammatory disease, and its incidence rate increases year by year with the progress of economic globalization. Currently, the diagnosis of IBD in child...
详细信息
ISBN:
(数字)9798350385724
ISBN:
(纸本)9798350385731
Inflammatory Bowel Disease (IBD) is a global chronic intestinal inflammatory disease, and its incidence rate increases year by year with the progress of economic globalization. Currently, the diagnosis of IBD in children mainly relies on endoscopic examination, but scoring endoscopic images is a challenging issue, especially in distinguishing different types of ulcers. To address this issue, this article designs a mobile application to accelerate data annotation processing and may provide reference for other unlabeled datasets. In the context of image segmentation, blurring labels has become an important issue. Deep learning methods are widely used in medical image segmentation, but their accuracy depends on high-quality annotated data. However, there are low-quality noise areas in the annotated data, and obtaining accurate and high-quality annotations becomes more time-consuming with limited annotation budgets. This article proposes a collaborative training framework to improve learning of noisy pixels. This framework determines the label confidence of an image by calculating the similarity between image pixels and surrounding pixels. Then, two parallel deep networks were constructed for semantic prediction, which aimed to guide each other on pixels that may have noise. By applying consistency in dual network prediction, the semantic information of uncertain pixels is corrected as much as possible. Experimental results have shown that this framework is slightly superior to models trained with pixel level precise labels, thus more effectively utilizing existing annotated data in the case of fuzzy labels.
In the field of efficient utility itemset mining, considering both internal and external utility values provides a more comprehensive approach compared to traditional frequency-based methods. However, the increased co...
详细信息
Purpose Compressed sensing (CS) reduces the measurement time of magnetic resonance (MR) imaging, where the use of regularizers or image priors are key techniques to boost reconstruction precision. The optimal prior ge...
详细信息
Purpose Compressed sensing (CS) reduces the measurement time of magnetic resonance (MR) imaging, where the use of regularizers or image priors are key techniques to boost reconstruction precision. The optimal prior generally depends on the subject and the hand-building of priors is hard. A methodology of combining priors to create a better one would be useful for various forms of imageprocessing that use image priors. methods We propose a theory, called prior ensemble learning (PEL), which combines many weak priors (not limited to images) efficiently and approximates the posterior mean (PM) estimate, which is Bayes optimal for minimizing the mean squared error (MSE). The way of combining priors is changed from that of an exponential family to a mixture family. We applied PEL to an undersampled (10%) multicoil MR image reconstruction task. Results We demonstrated that PEL could combine 136 image priors (norm-based priors such as total variation (TV) and wavelets with various regularization coefficient (RC) values) from only two training samples and that it was superior to the CS-SENSE-based method in terms of the MSE of the reconstructed image. The resulting combining weights were sparse (18% of the weak priors remained), as expected. Conclusion By the theory, the PM estimator was decomposed into the sparse weighted sum of each weak prior's PM estimator, and the exponential computational complexity for RCs was reduced to polynomial order w.r.t. the number of weak priors. PEL is feasible and effective for a practical MR image reconstruction task.
Backpropagation-based supervised learning has achieved great success in computer vision tasks. However, its biological plausibility is always controversial. Recently, the bioinspired Hebbian learning rule (HLR) has re...
详细信息
Backpropagation-based supervised learning has achieved great success in computer vision tasks. However, its biological plausibility is always controversial. Recently, the bioinspired Hebbian learning rule (HLR) has received extensive attention. Self-Organizing Map (SOM) uses the competitive HLR to establish connections between neurons, obtaining visual features in an unsupervised way. Although the representation of SOM neurons shows some brain-like characteristics, it is still quite different from the neuron representation in the human visual cortex. This paper proposes an improved SOM with multi-winner, multi-code, and local receptive field, named mlSOM. We observe that the neuron representation of mlSOM is similar to the human visual cortex. Furthermore, mlSOM shows a sparse distributed representation of objects, which has also been found in the human inferior temporal area. In addition, experiments show that mlSOM achieves better classification accuracy than the original SOM and other state-of-the-art HLR-based methods. The code is accessible at https://***/JiaHongZ/mlSOM.
暂无评论