Showing papers on "Layer (object-oriented design) published in 2013"

AU-aware Deep Networks for facial expression recognition

Abstract: In this paper, we propose to construct a deep architecture, AU-aware Deep Networks (AUDN), for facial expression recognition by elaborately utilizing the prior knowledge that the appearance variations caused by expression can be decomposed into a batch of local facial Action Units (AUs). The proposed AUDN is composed of three sequential modules: the first module consists of two layers, i.e., a convolution layer and a max-pooling layer, which aim to generate an over-complete representation encoding all expression-specific appearance variations over all possible locations; In the second module, an AU-aware receptive field layer is designed to search subsets of the over-complete representation, each of which aims at best simulating the combination of AUs; In the last module, multilayer Restricted Boltzmann Machines (RBM) are exploited to learn hierarchical features, which are then concatenated for final expression recognition. Experiments on three expression databases CK+, MMI and SFEW demonstrate the effectiveness of AUDN in both lab-controlled and wild environments. All our results are better than or at least competitive to the best known results.

Security system and method for operating systems

Abstract: A device comprising an operating system to run processes and a middleware layer operable to launch applications. An application launched by the middleware layer is run using one or more processes in the operating system. The operating system has a user layer and a kernel wherein the processes run in the user layer and interact with other processes running in the user layer through the kernel, the interaction being in response to calls to the kernel made by the processes. The device has one or more policy files defining policies for interaction of processes with the kernel of the device, and a monitor configured to monitor interaction of a process with the kernel to link or associate defined policies to the process, and to read code defined in the policy file or files linked or associated to the process.

Signaling change in output layer sets

Abstract: A system for decoding a video bitstream includes receiving a bitstream and a plurality of enhancement bitstreams together with receiving a video parameter set and a video parameter set extension. The system also receives an output layer set change message including information indicating a change in at least one output layer set.

Power demand side monitoring system and method

Abstract: The invention discloses a power demand side monitoring system and method. A cloud computing technology is used, power consumption information forms a large power information resource pool, the power load is accurately predicted, the energy-saving potential is accurately mined, and a powerful guarantee is provided for reasonable, ordered, high-efficiency and safe power utilization, so that the aim of continuously utilizing the power is fulfilled. The system comprises a user layer, a data layer, a service layer and a monitoring layer, wherein the user layer is used for dividing the energy-saving method of the whole distribution system according to the industry difference; the data layer is used for acquiring power user data of the user layer by employing an intelligent load control terminal; the service layer is used for processing the data acquired from the data layer by employing the cloud computing and big data technology; and the monitoring layer is connected to the cloud of the service layer and is used for monitoring a power load in real time, counting the power consumption, checking and managing the power service condition and checking the energy-saving potential, so that the service type of the energy-saving equipment is determined.

Three-layer Activity Recognition Combining Domain Knowledge and Meta-classification

Abstract: One of the essential tasks of healthcare and smart-living systems is to recognize the current activity of a particular user. Such activity recognition (AR) is demanding when only limited sensors are used, such as accelerometers. Given a small number of accelerometers, intelligent AR systems often use simple architectures, either general or specific for their AR. In this paper, a system for AR named TriLAR is presented. TriLAR has an AR-specific architecture consisting of three layers: (i) a bottom layer, where an arbitrary number of AR methods can be used to recognize the current activity; (ii) a middle layer, where the predictions from the bottom-layer methods are inputs for a hierarchical structure that combines domain knowledge and meta-classification; and (iii) a top layer, where a hidden Markov model is used to correct spurious transitions between the recognized activities from the middle layer. The middle layer has a hierarchical, three-level structure. First, a meta-classifier is used to make the initial separation between the most distinct activities. Second, domain knowledge in the form of rules is used to differentiate between the remaining activities, recognizing those of interest (i.e., static activities). Third, another meta-classifier deals with the remaining activities. In this way, each activity is recognized by the method best suited to it, leaving unrecognized activities to the next method. This architecture was tested on a dataset recorded using ten volunteers who acted out a complex, real-life scenario while wearing accelerometers placed on the chest, thigh, and ankle. The results show that TriLAR successfully recognized elementary activities using one or two sensors and significantly outperformed three standard, single-layer methods with all sensor placements.

Enterprise information system based on cloud computing

Abstract: The invention relates to the technical field of information, and in particular relates to an enterprise information system based on cloud computing. The system comprises a physical layer, a system layer, an application layer and a service layer, wherein the physical layer is used for providing infrastructural services for the system layer, the application layer and the service layer, virtualizing physical hardware resources and providing a uniform physical resource pool, and the physical resource pool comprises computing resources, storage resources, network resources and security resources; the system layer is positioned above the physical layer and used for providing system environment services for the application layer and the service layer; the application layer is positioned above the system layer and used for providing a uniform application resource pool and delivering services to provide application environment services, and the application environment services comprise distribution and management of application software, management of application data and configuration of application resources; and the service layer is positioned above the system layer and the application layer and used for providing service resource services and providing a uniform service application pool for enterprise service application according to a service logic.

Declarative Layer Composition with the JCop Programming Language

Abstract: Program behavior that relies on contextual information, such as physical location or network accessibility, is common in today's appli- cations, yet its representation at the source code level is not suciently supported by programming languages. With context-oriented program- ming (COP), context-dependent behavior can be explicitly modularized and dynamically activated. The available COP implementations oer lan- guage constructs that allow to describe context-dependent functionality and to specify for which control flows this functionality should be exe- cuted. Since these language constructs require modifications to the source code, the contemporary approaches limit the use of COP to program parts whose source code is accessible to the developer (the user code). The dy- namic control over context-dependent behavior in frameworks cannot be directly addressed by COP as this would imply changes to the source code. Instead, context-dependent behavior is addressed whenever a control flow from the framework code enters the user code. Context composition must be addressed at any of these control flow entry points, which may lead to a redundant specification of this functionality. As a result, dynamic control over layers emerges as a crosscutting concern that obstructs the separation of concerns. In this article, we discuss crosscutting layer composition in framework- based applications in detail. Moreover, we discuss limitations for the ex- pression of semantic relationships of layers that might lead to code du- plication. We present a framework-based application, a simple action ad- venture game that we implemented using a conventional COP language. Along this example, we show how our JCop language supports the decla- ration of layer composition and expression of layer relationships.

Intelligent skynet application platform

Abstract: The invention discloses an intelligent skynet application platform. Layered architecture is adopted by the intelligent skynet application platform, and standard general protocols and interfaces are used among layers. The intelligent skynet application platform comprises a collecting layer, a transmission layer, a core ability layer, an application layer and a client. The collecting layer is used for collecting videos, audios, pictures, data and alarms. The transmission layer provides high-quality physical transmission channels for all kinds of data flows and guarantees safe and reliable transmission of the data flows. The core ability layer achieves data access, data processing, system management and data interaction. A plurality of system management ability modules such as a data base module, a management service module, an intelligent analysis module, a streaming media distributing module, a comprehensive assessment module and an alarm and positioning module are arranged in the system, cooperatively work, and can automatically conduct processes such as data access, data processing, system management and data interaction so as to achieve automatic and intelligent monitoring. The intelligent skynet application platform can be combined with GIS or PGIS positioning so as to achieve fast positioning of an event.

Method and apparatus for scalable encoding and decoding

Abstract: PURPOSE: A scalable video coding and decoding method and a device using the same are provided to supply various video services on a time, a space, or an image quality side according to a use purpose or various demands. CONSTITUTION: An encoding device and a decoding device determine whether inter layer prediction is applied to an enhanced layer(S1610). If the inter layer prediction is applied, the encoding device and the decoding device perform the prediction of a current layer of the enhanced layer based on reference information generated from a decoded reference picture of a reference layer(S1630). The encoding device and the decoding device restore or encode the current picture of the enhanced layer(S1640). [Reference numerals] (AA) Start; (BB) No; (CC) Yes; (DD) End; (S1610) Predicting an inter layer?; (S1620) Predicting in an enhanced layer; (S1630) Predicting based on a generalized reference picture or a generalized reference block generated from information of a reference layer; (S1640) Restoring a picture of the enhanced layer

Method and system for managing background operations in a multi-layer memory

Abstract: A multi-layer memory and method for performing background maintenance operations are disclosed. The memory includes a plurality of flash memory die having multiple layers, where each layer is made up of flash memory cells having a greater bit per cell storage capacity than then prior layer and each layer may have a plurality of partitions for different data types. A controller managing the flash memory die is configured to identify an idle die and determine if a layer in the die satisfies a background maintenance criterion. Upon identifying a layer satisfying the background maintenance criterion, the valid data from reclaim blocks in the layer is relocated into a relocation block in the same layer until the relocation block is filled and the background maintenance cycle ends.

Multi-objective energy management methods for micro-grids

Abstract: Systems and methods are disclosed for multi-objective energy management of micro-grids. A two-layer control method is used. In the first layer which is the advisory layer, a Model Predictive Control (MPC) method is used as a long term scheduler. The result of this layer will be used as optimality constraints in the second layer. In the second layer, a real-time controller guarantees a second-by-second balance between supply and demand subject to the constraints provided by the advisory layer.

Content management using multiple abstraction layers

Abstract: Systems, devices, methods and computer program products improve various aspects of a content management system. In one scenario, one or more contents are accessed at a content handling device. The content handling device is configured to operate using multiple abstraction layers including an application layer, an operating system layer, a firmware layer and a hardware layer. Content management operations are conducted on the one or more contents using at least two abstraction layers. Such content management operations include conducting watermark extraction and content screening operations at a first layer for a first watermark message having a first value, conducting watermark extraction and content screening operations at a second layer for a second watermark message having a second value, and initiating one or more enforcement actions corresponding to the first value and the second value based at least in-part on respective results of the content screening.

Data integrated management scheme for Internet of Things based on middleware framework

Abstract: The invention discloses a data integrated management scheme for an Internet of Things based on a middleware framework. An acquisition and integrated management framework for the data in the Internet of Things is divided into four layers through the data integrated management scheme, and the four layers are respectively a user interface layer, an application layer, a data integration layer and a bottom isomerous layer. In an overall environment of the Internet of Things, aiming at the data integration layer, the middleware framework of the data acquisition in the Internet of Things is provided, and the functional modules of the middleware and specific design flows of the functional modules are elaborated. Based on the middleware, the integrated management frame of an isomerous database is provided so that the middleware framework and the integrated management framework are combined together, the unified management of the data in the Internet of Things can be realized, and the requirements of users can be met well. Therefore, the data integrated management scheme has a certain actual application value.

Public bicycle internet of things system

Abstract: The invention discloses a public bicycle internet of things system, comprising a perception layer, a centralized control layer, a data processing layer, a system service layer and a system management layer. The perception layer is used for acquiring public bicycle equipment and citizen bicycle lease IC card information; the centralized control layer finishes the network building, the information receiving and transmitting of the perception layer, and the data conversion processing of the information; the data processing layer is used for carrying out creating, management, data classification and processing on a system database; the system service layer is used for processing all application operations related to users of a public bicycle system; and the system management layer is used for processing all operations related to management personnel and system maintenance personnel of the public bicycle system. According to the public bicycle internet of things system, the system scale can be configured dynamically, a distributed processing strategy is adopted for system data processing, a bicycle parking locking column and a bicycle branch management box have independent offline processing capability, so that the communication pressure on a system background is greatly reduced, and the system reliability, extensibility, applicability and fault-tolerant ability are greatly improved.

Device and design method for preventing robot from generating accidental collision

Abstract: The invention relates to a device for preventing a robot from generating accidental collision. The device comprises a robot body, a controller and a safety protective device; the controller is connected to the robot body and is provided with a scram button; a viscoelasticity buffering layer is attached to the robot body and is attached to one or more touch sensors; the touch sensors are connected with the input interface of the safety protective device; and the scram button is connected to the output interface of the safety protective device. The buffering layer is wrapped on a connection rod arm of the robot body for buffering and reducing the impact force generated during colliding, and the scram of the robot is realized by feedback of the touch sensors in the effective compression stroke of the buffering layer, so a collided object is protected, the severity of accident collision generated by the robot can be reduced under the premise that the existing robot does not need to be greatly changed, the realization method is simple and the reliability is high.

Computer system and control method

Abstract: This computer system is configured from a server (200), which is provided with a flash memory drive (204) for caching, a storage (260), which is provided with a storage hierarchy composed of an SSD (267) and a HDD (268), and a management server, which is provided with a page hierarchical layer determination program (503) for determining a storage hierarchical layer on which data is to be stored. The page hierarchical layer determination program (503) transfers, to the storage hierarchical layer formed by the HDD (268), data which is selected from among pages having a high caching rate to the flash memory drive (204) and which has a high read access rate, thereby preventing the same data from being stored in the flash memory drive (204) and the SSD (267) simultaneously. Further, data having a relatively high write access rate is transferred to the storage hierarchical layer formed by the SSD (267) so as to prevent reductions in write processing performance.

Electrical power system three-dimensional digitalization operation platform

Abstract: An electrical power system three-dimensional digitalization operation platform comprises a data service layer, a function logic layer, a management unit, a basic unit, a public unit, a presentation layer, and an application layer. The data service layer comprises a three-dimensional model data bank, a topography image data bank, a business data bank, and a system management bank, wherein the business data bank and the system management bank store abutting joint relations with each power grid business platform. The function logic layer comprises a power grid business unit which is communicated with the data service layer, and is connected with the basic unit and the management unit, wherein the power grid business unit supplies six power grid business components, namely an auxiliary plan component, an auxiliary design component, an auxiliary operation-maintenance component, a construction progress component, an auxiliary emergency dealing component, and a data adjusting component. The management unit provides a system management component and a system maintenance component. The basic unit provides a basic three-dimensional browsing component and a three-dimensional analysis component which are both relied by a three-dimensional integrated platform. The public unit provides a main frame, a plug-in engine component, and a data access component which are all of the system. The presentation layer supplies a three-dimensional scene picture, a picture layer management view, a business data display view, an auxiliary design view for building a three-dimensional power grid. The application layer sends an analysis command of a leader and a maintain demand of data bank administrators to the data server layer, the function logic layer, and the application layer.

Scalable Distributed Two-Layer Data Structures SD2DS

Abstract: Scalability and fault tolerance are important features of modern applications designed for the distributed, loosely-coupled computer systems. In the paper, two-layer scalable structures for storing data in a distributed RAM of a multicomputer SD2DS are introduced. A data unit of SD2DS a component is split into a header and a body. The header identifies the body and contains its address in a network. The headers are stored in the first layer of SD2DS, called the component file, while the bodies are stored in the second layer, called the component storage. Both layers are managed independently. Details of the management algorithms are given, along with SD2DS variant suitable for storing plain records of data. The SD2DS is compared to similar distributed structures and frameworks. Comparison considerations together with test results are also given. The results proved superiority of SD2DS over similar structures.

Systems and methods for indirectly associating logical and physical display content

Abstract: Systems and methods include utilizing a layer stack to indirectly associate logical and physical display content. A layer stack may decouple content from content presentation details at a physical display, facilitating the implementation of mirroring, spanning, and other multiple-display modes across non-contiguous display devices with disparate resolutions, densities, and other characteristics, while maintaining native device configuration settings. In one implementation, a layer stack may be a collection of surfaces. The layer stack may be associated with a first logical display having a first resolution. A region containing parts of one or more surfaces, at a first position of the layer stack and corresponding to the first resolution of the first logical display, may be rendered and output, based on a display projection, to a first physical display. Further implementations may use combinations of additional logical displays, physical displays, or layer stacks to implement various multiple-display modes.

Managing a set of software modules in a software runtime environment

Abstract: A method, system, and/or computer program product manages a set of software modules in a software runtime environment, where the set of software modules has one or more dynamic dependencies and one or more static dependencies. Each software module having a static dependency on another software module in a set of software modules is identified. Each subset of the set of software modules that is interdependent with another subset due to static dependencies between subsets is identified. In response to a change in state of a module in an identified subset, the change in state is applied to all modules in the identified subset.

Multi layer display apparatus

Abstract: A multilayer display apparatus for outputting a single integrated image includes a plurality of layer displays to represent an image, and an image control unit to control a single integrated image to be output through overlapping images represented by the plurality of layer displays.

Security system and method for the android operating system

Abstract: A method of linking a security policy stored in a policy database that is specific to an application in the application layer with a new corresponding process launched in the Linux layer in a security system for an operating system running on a device that comprises a Linux-based kernel. The system architecture is defined by a middleware layer between the Linux layer associated with the kernel and the higher application layer comprising the applications.

Multi-level modeling as a society of interacting models

Abstract: We use the multi-agent paradigm to consider multi-level modeling as a society of interacting models.We propose a framework to better specify the concepts used in multi-level modeling and their relationships. This framework is implemented through the AA4MM metamodel, which benefits from a middleware layer. We use it to present some proofs of concept examples where we show the ability of this approach to rapidly change from one pattern of interaction to another by reusing some of the components.

Layer dependency and priority signaling design for scalable video coding

Abstract: Signaling of layer dependency and/or priority of dependent layers in a video parameter set (VPS) may be used to indicate the relationship between an enhancement layer and its dependent layers, and/or prioritize the order of the dependent layers for multiple layer scalable video coding of HEVC for inter-layer prediction. A method may include receiving a bit stream that includes a video parameter set (VPS). The VPS may include a dependent layer parameter that indicates a dependent layer for an enhancement layer of the bit stream. The dependent layer parameter may indicate a layer identification (ID) of the dependent layer. The VPS may indicate a total number of dependent layers for the enhancement layer. The VPS may include a maximum number of layers parameter that indicates a total number of layers of the bit stream. The total number of dependent layers for the enhancement layer may not include the enhancement layer.

Automatic generation of different attribute values for detecting a same type of web application layer attack

Abstract: According to one embodiment, a computing device is coupled to a web application layer attack detector (AD), which itself is coupled between an HTTP client and a web application server. The computing device automatically learns a new condition to detect a first type of web application layer attack. Responsive receiving a web application layer message from the HTTP client that violates a rule for detecting the first type of web application layer attack, the AD transmits an alert package to the computing device, which uses the alert package, and optionally other alert packages, to automatically generate a new set of attribute values for each of a set of attribute identifiers to be transmitted to the AD or optionally other ADs for use in a different rule than the violated rule. The different rule is another attack specific rule for detecting the first type of web application layer attack.

Cloud search platform

Abstract: The invention discloses a cloud search platform. The cloud search platform comprises a user authority system and a second presentation layer. The cloud search platform is characterized by further comprising a basic data layer, a data persistence layer, a JAVA application support layer and a first persistence layer which sequentially perform bidirectional data communication, wherein the user authority system performs bidirectional data communication with the basic data layer, the data persistence layer, the JAVA application support layer and the first persistence layer respectively; and the second persistence layer performs bidirectional data communication with the basic data layer. The cloud search platform provided by the invention has the characteristics of all-directional search, high reliability, high memory space, low delay and the like; and a cloud search platform which is more advanced, easy to use, and fast in response and analysis speeds is provided for enterprises or governments.

Modularized multi-machine parallel-connection large-power APF (active power filter) control system and realization method

Abstract: The invention discloses a modularized multi-machine parallel-connection large-power APF (active power filter) control system and a realization method. By coordinating an upper layer controller and a lower layer controller, an effect for reasonably allocating the compensative capacity and improving the compensation precision can be realized; the upper layer controller is used for carrying out the coordination and switching control on each power module, for example, the upper layer controller is used for operating and monitoring the quantity of the power modules to be put into use according to a load requirement, numbering the power modules according to a rule and eliminating the switching subharmonic by utilizing the carrier phase shifting; the lower layer controller is used for receiving a compensation order of the upper layer controller and a synchronous signal of a previous sequence number controller and feeding back state information of the power module to the upper layer controller. By adopting the modularized multi-machine parallel-connection large-power APF control system, the compensation capacity of the power module and the computation of an upper layer and a lower layer of the control system are reasonably allocated, the entire system can be enabled to run stably and orderly, the stopping of the entire system caused by the malfunction of one power module can be avoided, and the feasibility and practical value are good.

Artificial intelligent type hydraulic support electric-hydraulic control system

Abstract: The invention discloses an artificial intelligent type hydraulic support electric-hydraulic control system which comprises a hardware layer, a system layer, a specialist system and a data base. The system conducts communication through a communication bus and adopts a full duplex signal. The system layer comprises management layer Agents and control layer Agents, and each hydraulic support corresponds to a control layer Agent. The specialist system is connected with a communication bus signal, and the data base stores sensing parameters of three machines and is in signal connection with the communication bus. The artificial intelligent type hydraulic support electric-hydraulic control system has the advantages that the system layer calls the current sensing parameters of the three machines saved by the data base, ultra long instruction words are formed finally by calling the sensing parameters of the three machines through the specialist system, the sensing parameters are transmitted to each hydraulic support controller through the communication bus, the control layer Agents make an automatic movement decision on data information, the management layer Agents coordinate movements of the control layer Agents, and intellectualization of the hydraulic support electric-hydraulic control system of a whole fully mechanized coal mining face is achieved.