ID 原文 译文
40946 所以视网膜血管树的精确分割己成为计算机辅助诊断的先决条件。 Thus, accurate segmentation of the retinal vascular tree has become a prerequisite for computer-aided diagnosis.
40947 随着卷积神经网络在医学图像分割中的应用，一些分割性能优越的网络逐渐被提出。 With the application of convolutional neural networks in medical image segmentation, some Network with optimized segmentation performance have been gradually proposed.
40948 但是他们忽略了上下文多视野的关注，导致微血管分支很难分出。 However, these methods ignore the attention of context and multi-view, making it difficult to separate microvascular branches.
40949 为了解决此问题，本文提出了一种多视野上下文关注的网络架构。 To solve this problem, a multi-view context attention network architecture is proposed.
40950 该网络融入一个新的多视野关注模块，该模块能够在扩大感受野的同时关注有效信息，防止微血管分割断裂。 Firstly, the network adds a new multi-view attention module, which can effectively focus on the valid information while expanding the receptive field, to prevent the fracture of microvascular segmentation.
40951 其次网络融入注意力门控机制，将低级特征和高级特征相结合，产生更具代表性的新特征，进一步提升微血管分割的性能。 Secondly, the network integrates attention gate, it combines low-level features with high-level features, produced more representative new features, and further improve the performance of microvascular segmentation.
40952 在公开的两个眼底数据集DRIVE和CHASE DB1进行了评估，实验结果显示在准确率，灵敏度，交并比和AUC(ROC曲线下的面积)上均优于目前的新算法。 It is evaluated on two public fundus datasets DRIVE and CHASE DB1, the experimental results show that the algorithm is superior to the state-of-the-art methods in accuracy, sensitivity, intersection-over-union and AUC(area under ROC curve).
40953 针对风电功率数据的时序性特点，提出降噪循环神经网络模型对电场中短期内的风电功率进行预测。 Aiming at time series characteristics of wind power data, the Denoising Recurrent Neural Network model is proposed to predict the wind power in the short and medium term of the electric field.
40954 通过模型能够挖掘其蕴含的知识，提高电力系统的稳定性，优化电力调度。 Through the model, the knowledge contained in it can be mined to improve the stability of the power system and optimize the power dispatching.
40955 模型首先采用循环神经网络构建一个编码-解码结构，设计编码器从序列变量中获取相应的深度特征，再通过解码器对深度特征进行解码，还原输入序列的状态并预测下一时刻的输出。 An encoding-decoding structure is first designed in the recurrent neural network model, and the encoder is designed to obtain corresponding depth features from the sequence variables, and then the decoder decodes the depth features, restore the state of the input sequence and makes a prediction.