Automatic classification of Human Epithelial Type-2 (HEp-2) specimen patterns is an important yet challenging problem in medical image analysis. Most prior works have primarily focused on cells images classification p...
详细信息
ISBN:
(纸本)9781509048489
Automatic classification of Human Epithelial Type-2 (HEp-2) specimen patterns is an important yet challenging problem in medical image analysis. Most prior works have primarily focused on cells images classification problem which is one of the early essential steps in the system pipeline, while less attention has been paid to the classification of whole-specimen ones. In this work, a specimen pattern recognition system combining convolutional neural networks (CNNs) and pattern histogram was proposed. The pattern histograms were obtained based on the prediction of each single cell inside the specimens. Two strategies were designed to predicted the pattern of a whole specimen: 1) the most dominant cell pattern in pattern histogram was represented as the specimen pattern, 2) the pattern histograms were employed as bags of patterns and then were trained and predicted separately by a SVM classifier. Experimental results show that the proposed system is effective and achieves high classification accuracy on public benchmark datasets. We further evaluate the robustness of the proposed framework by testing trained CNNs on another different dataset, demonstrating that the system is robust to inter-lab data.
Classic OPC, released by OPC Foundation, is well accepted and applied in industrial automation, which led to many OPC products on the market from a variety of companies. However, OPC technology was based on retiring M...
Classic OPC, released by OPC Foundation, is well accepted and applied in industrial automation, which led to many OPC products on the market from a variety of companies. However, OPC technology was based on retiring Microsoft COM/DCOM. The OPC Unified Architecture was introduced as the new generation specification with the main goal of keeping all the functionality of Classic OPC and switching from COM/DCOM technology to state -of-the-art web services. The OPC Foundation has been also developing OPC UA Toolkit that provide a collection of libraries, classes, and interfaces which make developers and programmers easy to create and implement OPC UA components. However, this toolkit is insufficient for developers and programmers to implement real monitoring and control applications from industry due to the limitations of such a toolkit, the complexity of related decision tasks and information systems, etc. In this paper, an OPC UA (Unified Architecture) client framework is proposed and developed by using OPC UA specifications, Service Oriented Architecture (SOA), web services, XML, OPC UA SDK, etc. This framework minimizes the efforts of developers and programmers in learning new techniques and allows system arc hitects and designers to perform dependency analysis on the development of monitoring and control applications. The initial results from the system implemented by Visual Studio 2008 are also provided.
The paper proposes a novel method to detect fire and flame by processing the video data captured by an ordinary camera monitoring an open scene, which combined fire flicker and color clues to reach a final decision. F...
详细信息
The paper proposes a novel method to detect fire and flame by processing the video data captured by an ordinary camera monitoring an open scene, which combined fire flicker and color clues to reach a final decision. First the image difference method and color analysis technique based on the HIS color model are applied to extract the fire-like moving region between two consecutive frames. Then the spectrum of the fire-like region flickering could be got by the Discrete Fourier Transform of the fire-like moving region sequence. Finally, a simple method is devised to estimate the fire alarm grade so that users could be informed with a proper alarm. Experimental results show that the fire detecting algorithm is fast, effective, and robust.
The concept of a "Networked Physical World" originated from the Auto-ID Center, now called the Auto-ID Labs. Such a system can be realized with a combination of automatic identification technology and a ubiq...
详细信息
The concept of a "Networked Physical World" originated from the Auto-ID Center, now called the Auto-ID Labs. Such a system can be realized with a combination of automatic identification technology and a ubiq...
详细信息
The concept of a "Networked Physical World" originated from the Auto-ID Center, now called the Auto-ID Labs. Such a system can be realized with a combination of automatic identification technology and a ubiquitous computer network that will glue the physical world together. The ability to form a ubiquitous network of physical objects has a wide range of applications including manufacturing automation, supply chain management and collection of sensor derived data We describe the building block system components of a distributed ubiquitous RFID network aimed at enabling ubiquitous sensing with RFID and explore the data flows within the system.
暂无评论