With the accelerating process of social informatization,our personal information security and Internet sites,etc.,have been facing a series of threats and ***,well-developed neuralnetwork has seen great advancement i...
详细信息
With the accelerating process of social informatization,our personal information security and Internet sites,etc.,have been facing a series of threats and ***,well-developed neuralnetwork has seen great advancement in natural language processing and computer vision,which is also adopted in intrusion *** this research,a hybrid model integrating multi-scale convolutional neural network and Long Short-term Memory network(MSCNN-LSTM)is designed to conduct the intrusion ***-scaleconvolutionalneuralnetwork(MSCNN)is used to extract the spatial characteristics of data *** Long Short-term Memory network(LSTM)is responsible for processing the temporal *** data set used in this experiment is KDDCUP99 with different probability distributions in the training set and test set involving some newly emerging attack types,making the data more *** a result,this type of data set is widely applied in the simulation experiment of intrusion *** this experiment,the assessment indices such as the accuracy rate,recall rate and F1 score are introduced to check the performance of this model.
Lung cancer is a global and dangerous disease, and its early detection is crucial for reducing the risks of mortality. In this regard, it has been of great interest in developing a computer-aided system for pulmonary ...
详细信息
Lung cancer is a global and dangerous disease, and its early detection is crucial for reducing the risks of mortality. In this regard, it has been of great interest in developing a computer-aided system for pulmonary nodules detection as early as possible on thoracic CT scans. In general, a nodule detection system involves two steps: (i) candidate nodule detection at a high sensitivity, which captures many false positives and (ii) false positive reduction from candidates. However, due to the high variation of nodule morphological characteristics and the possibility of mistaking them for neighboring organs, candidate nodule detection remains a challenge. In this study, we propose a novel multi-scale Gradual Integration convolutionalneuralnetwork (MGI-CNN), designed with three main strategies: (1) to use multi-scale inputs with different levels of contextual information, (2) to use abstract information inherent in different input scales with gradual integration, and (3) to learn multi-stream feature integration in an end-to-end manner. To verify the efficacy of the proposed network, we conducted exhaustive experiments on the LUNA16 challenge datasets by comparing the performance of the proposed method with state-of-the-art methods in the literature. On two candidate subsets of the LUNA16 dataset, i.e., V1 and V2, our method achieved an average CPM of 0.908 (V1) and 0.942 (V2), outperforming comparable methods by a large margin. Our MGI-CNN is implemented in Python using TensorFlow and the source code is available from https://***/ku-milab/MGICNN. (C) 2019 Elsevier Ltd. All rights reserved.
作者：
Zeng, DanZhang, ShunChen, FanshengWang, YuemingShanghai Univ
Key Lab Specialty Fiber Opt & Opt Access Networks Joint Int Res Lab Specialty Fiber Opt & Adv Commu Shanghai Inst Adv Commun & Data Sci Shanghai 200444 Peoples R China Chinese Acad Sci
Key Lab Intelligent Infrared Percept Shanghai 200083 Peoples R China Chinese Acad Sci
Key Lab Space Act Optoelect Technol Shanghai Inst Tech Phys Shanghai 200083 Peoples R China
Garbage detection is important for environmental monitoring in large areas. However, the manual patrol is time-consuming and labor-intensive. This paper proposes a method for monitoring garbage distribution in large a...
详细信息
Garbage detection is important for environmental monitoring in large areas. However, the manual patrol is time-consuming and labor-intensive. This paper proposes a method for monitoring garbage distribution in large areas with airborne hyperspectral data. Since there is no public hyperspectral garbage dataset, a hyperspectral garbage dataset Shandong Suburb Garbage is labeled and published. For garbage detection, a new hyperspectral image (HSI) classification network MSCNN (multi-scale convolutional neural network) is proposed to classify the pixels of HSI data and generate binary garbage segmentation map. Unsupervised region proposal generation algorithm Selective Search and None Maximum Suppression (NMS) are used to extract the location and the size of garbage areas based on the garbage segmentation map. The experiment results show that the proposed algorithm has a good performance on garbage detection in large areas. In addition, the MSCNN has achieved better performance in comparison with other HSI classification methods in the public HSI datasets Indian Pines and Pavia University.
Fabric defect detection plays an important role in controlling the quality of textile production. In this article, a novel fabric defect detection algorithm is proposed based on a multi-scaleconvolutionalneural netw...
详细信息
Fabric defect detection plays an important role in controlling the quality of textile production. In this article, a novel fabric defect detection algorithm is proposed based on a multi-scale convolutional neural network and low-rank decomposition model. First, multi-scale convolutional neural network, which can extract the multi-scale deep feature of the image using multiple nonlinear transformations, is adopted to improve the characterization ability of fabric images with complex textures. The effective feature extraction makes the background lie in a low-rank subspace, and a sparse defect deviates from the low-rank subspace. Then, the low-rank decomposition model is constructed to decompose the feature matrix into the low-rank part (background) and the sparse part (salient defect). Finally, the saliency maps generated by the sparse matrix are segmented based on an improved optimal threshold to locate the fabric defect regions. Experimental results indicate that the feature extracted by the multi-scale convolutional neural network is more suitable for characterizing the fabric texture than the traditional hand-crafted feature extraction methods, such as histogram of oriented gradient, local binary pattern, and Gabor. The adopted low-rank decomposition model can effectively separate the defects from the background. Moreover, the proposed method is superior to state-of-the-art methods in terms of its adaptability and detection efficiency.
暂无评论