In this paper, a SoC system with ARM processor and convolution accelerator is designed for cnn algorithms on the ZC706 evaluation board. Using tiling technology and loop reorganization, the system has a high data reus...
详细信息
ISBN:
(纸本)9781728102863
In this paper, a SoC system with ARM processor and convolution accelerator is designed for cnn algorithms on the ZC706 evaluation board. Using tiling technology and loop reorganization, the system has a high data reuse rate, thus greatly reducing the data bandwidth between the on-chip buffer and DDR memory. This convolution accelerator supports different kernel size from 1x1 to 11x11, while the activation functions supported are ReLU and Leaky ReLU. The processor of the SoC is mainly responsible for controlling and processing other computations of the cnn, such as LRN and pooling, which makes the system more versatile and flexible. At the working frequency of 100MHz, the peak performance can reach 45.16 GFLOPS, which is 142.8x faster than Cortex-A9 and the energy efficiency is 219.5x better compared to i7-4790K.
Computer network security is an important issue faced in today's Internet era, and traditional security techniques can no longer meet the complex and changing network attacks and threats. In this paper, based on t...
详细信息
ISBN:
(纸本)9798400716959
Computer network security is an important issue faced in today's Internet era, and traditional security techniques can no longer meet the complex and changing network attacks and threats. In this paper, based on the convolutional neural network (cnn) algorithm, computer network security technology is studied. The detection and prevention of network attacks are realized by deep learning and classification of network traffic data. By constructing a cnn model, extracting the features of network traffic data and classifying them, malicious traffic can be effectively identified and corresponding defense measures can be taken in time. The experimental results show that the computer network security technology based on cnn algorithm has high accuracy and reliability, and can effectively improve the security of the network.
Deep learning has become a crucial instrument for medical research in recent years. Computer science-based mathematics have been extensively used in research to identify and forecast various diseases. This research pr...
详细信息
Deep learning has become a crucial instrument for medical research in recent years. Computer science-based mathematics have been extensively used in research to identify and forecast various diseases. This research presents a new model called FEM-Dcnn that integrates the Finite Element Method (FEM), Deep Auto Encoder Algorithm (DAE), and Convolutional Neural Network (cnn) techniques. The performance of two deep learning (DL) models was combined rather than just one to enable the execution and prediction of the outcome with greater accuracy. The LIDC-IDRI dataset, which is publicly available, was used in this study;the dataset comprises a CT scan along with annotations that enhance the understanding of the data as well as information pertaining to each CT scan. In this work, an ensemble approach has been developed for solving the issue of lung nodule detection and thus coming up with a robust automated model for lung cancer diagnosis. The objective of DAE is to extract the features of various objects in CT-scan images. The extracted features are then used to build the cnn network. This combination is aimed at precisely determining the boundaries of different objects, allowing for effective image segmentation. The use of FEM helps to decrease computational complexity when integrating DAE and cnn, thereby achieving the objective of this study. The proposed approach in this study outperformed the single cnn algorithms based on the employed performance metrics.
One of the key factors that determines the loss of agricultural production and in its yield is the discernment and recognition of plant diseases. Plant disease research is the investigation of any visible points of de...
详细信息
作者：
Xie, BinHe, XiaoyuLi, YiCent S Univ
Sch Informat Sci & Engn Changsha Hunan Peoples R China Cent S Univ
Xiangya Hosp China Mobile Joint Lab Mobile Hlth Minist Educ Changsha Hunan Peoples R China
In the area of human-computer interaction (HCI) and computer vision, gesture recognition has always been a research hotspot. With the appearance of depth camera, gesture recognition using RGB-D camera has gradually be...
详细信息
In the area of human-computer interaction (HCI) and computer vision, gesture recognition has always been a research hotspot. With the appearance of depth camera, gesture recognition using RGB-D camera has gradually become mainstream in this field. However, how to effectively use depth information to construct a robust gesture recognition system is still a problem. In this paper, an RGB-D static gesture recognition method based on fine-tuning Inception V3 is proposed, which can eliminate the steps of gesture segmentation and feature extraction in traditional algorithms. Compared with general cnn algorithms, the authors adopt a two-stage training strategy to fine-tune the model. This method sets a feature concatenate layer of RGB and depth images in the cnn structure, using depth information to promote the performance of gesture recognition. Finally, on the American Sign Language (ASL) Recognition dataset, the authors compared their method with other traditional machine learning methods, cnn algorithms, and the RGB input only method. Among three groups of comparative experiments, the authors' method reached the highest accuracy of 91.35%, reaching the state-of-the-art currently on ASL dataset.
Automated deep learning and data mining algorithms can provide accurate detection, frequency patterns, and predictions of dangerous goods passing through motorways and tunnels. This paper presents a post-processing im...
详细信息
Automated deep learning and data mining algorithms can provide accurate detection, frequency patterns, and predictions of dangerous goods passing through motorways and tunnels. This paper presents a post-processing image detection application and a three-stage deep learning detection algorithm that identifies and records dangerous goods' passage through motorways and tunnels. This tool receives low-resolution input from toll camera images and offers timely information on vehicles carrying dangerous goods. According to the authors' experimentation, the mean accuracy achieved by stage 2 of the proposed algorithm in identifying the ADR plates is close to 96% and 92% of both stages 1 and 2 of the algorithm. In addition, for the successful optical character recognition of the ADR numbers, the algorithm's stage 3 mean accuracy is between 90 and 97%, and overall successful detection and Optical Character Recognition accuracy are close to 94%. Regarding execution time, the proposed algorithm can achieve real-time detection capabilities by processing one image in less than 2.69 s.
暂无评论