this paper presents a novel approach for behavior recognition from video data. A biologically inspired action representation is derived by applying a clustering algorithm to sequences of motion images. To obey the tem...
详细信息
ISBN:
(纸本)9783540742715
this paper presents a novel approach for behavior recognition from video data. A biologically inspired action representation is derived by applying a clustering algorithm to sequences of motion images. To obey the temporal context, we express behaviors as sequences of n-grams of basic actions. Novel video sequences are classified by comparing histograms of action n-grams to stored histograms of known behaviors. Experimental validation shows a high accuracy in behavior recognition.
We have theoretically shown that multiferroic nanomagnets (consisting of a piezoelectric and a magnetostrictive layer) could be used to perform computing while dissipating similar to few 100 kT/bit at clock rates of s...
详细信息
ISBN:
(纸本)9781467322003
We have theoretically shown that multiferroic nanomagnets (consisting of a piezoelectric and a magnetostrictive layer) could be used to perform computing while dissipating similar to few 100 kT/bit at clock rates of similar to 1GHz [1,2,3]. they can act as memory elements [2], binary logic gates [3, 4] and associative memory for four state logic [5, 6]. the latter enables signal processing functions such as ultrafast image reconstruction and patternrecognition [7]. this talk will provide an overview of our research in modeling stress induced nanoscale magnetization dynamics, its application to ultra low energy hybrid spintronic/straintronics memory and information processing, and discuss preliminary experimental work in fabrication and experimental demonstration of these devices.
there are four main problems that limit application of patternrecognition techniques for recognition of abnormal cardiac left ventricle (LV) wall motion: 1) Normalization of the LV's size, shape, intensity level ...
详细信息
ISBN:
(纸本)9783642042706
there are four main problems that limit application of patternrecognition techniques for recognition of abnormal cardiac left ventricle (LV) wall motion: 1) Normalization of the LV's size, shape, intensity level and position;2) defining a spatial correspondence between phases and Subjects;3) extracting features;4) and discriminating abnormal from normal wall motion. Solving these four problems is required for application of patternrecognition techniques to classify the normal and abnormal LV wall motion. In this work, we introduce a normalization scheme to solve the first and second problems. Withthis scheme, LVs are normalized to the same position, size, and intensity level. Using the normalized images, we proposed an intra-segment classification criterion based on a con-elation measure to solve the third and fourth problems. Application of the method to recognition of abnormal cardiac MR LV wall motion showed promising results.
Scientific evidence has shown that long-term sedentary behaviour is detrimental to human health. therefore, a trend appears in the field of healthy lifestyle promotion that more attention is drawn to sedentary behavio...
详细信息
ISBN:
(纸本)9781450364508
Scientific evidence has shown that long-term sedentary behaviour is detrimental to human health. therefore, a trend appears in the field of healthy lifestyle promotion that more attention is drawn to sedentary behaviour rather than only physical activity. However, technology-based mobile health intervention tools targeting reducing sedentary behaviour are still lacking. this paper aims to explore a solution for sedentary behaviour change through supporting action planning. Action planning can not only bridge the intention-behavior gap in controlled motivation processes, but also enforce the cue-behavior association in unconscious processes. We present a smartphone-based personal mobility pattern visualization, with which we expect the users can make better action plans. the interactive visualization integrates temporal and spatial patterns of personal sedentary and walking behaviour, to provide explicit hints on when, where, and how to reduce sedentary behaviour and increase daily steps. We also present our experimental design to evaluate the visualization-based intervention tool.
this paper proposes a novel biometric authentication method based on the recognition of drivers' dynamic handgrip on steering wheel. A pressure sensitive mat mounted on a steering wheel is employed to collect hand...
详细信息
this paper discusses a prototype of a temporal pattern predictor, which was built on specifications derived from the descriptions of the "Ergotrix" temporal memory network in Valentino Braitenberg s "Ve...
详细信息
this paper presents ontlology-based architecture for patternrecognition in the context of static source code analysis. the proposed system has three subsystems: parser, OWL ontologies and analyser. the parser subsyst...
详细信息
ISBN:
(纸本)9783540855620
this paper presents ontlology-based architecture for patternrecognition in the context of static source code analysis. the proposed system has three subsystems: parser, OWL ontologies and analyser. the parser subsystem translates the input coded to AST that is constructed as an XML tree. the OWL ontologies define code patterns and general programming concepts. the analyser subsystem constructs instances of the input code as ontology individuals and asks the reasoner to classify them. the experience gained in the implementation of the proposed system and some practical issues are discussed. the recognition system successfully integrates the knowledge representation field and static code analysis. resulting in greater flexibility of the recognition system.
the proceedings contain 232 papers. the topics discussed include: convolutional neural network-based approach for citrus diseases recognition;using temporal conceptual graphs and neural networks for big data-based att...
ISBN:
(纸本)9781728143286
the proceedings contain 232 papers. the topics discussed include: convolutional neural network-based approach for citrus diseases recognition;using temporal conceptual graphs and neural networks for big data-based attack scenarios reconstruction;taxi demand prediction with LSTM-based combination model;counting attention based on classification confidence for visual question answering;an optimization method of WebP images lossy compression algorithm for FPGAs;self-adaptive address mapping mechanism for access pattern awareness on DRAM;a high-performance self-learning antelopes migration algorithm (SAMA) for global optimization;and using resource use data and system logs for HPC system error propagation and recovery diagnosis.
the unsupervised learning of spectro-temporal speech patterns is relevant in a broad range of tasks. Convolutive non-negative matrix factorization (CNMF) and its sparse version, convolutive non-negative sparse coding ...
详细信息
ISBN:
(纸本)9781618392701
the unsupervised learning of spectro-temporal speech patterns is relevant in a broad range of tasks. Convolutive non-negative matrix factorization (CNMF) and its sparse version, convolutive non-negative sparse coding (CNSC), are powerful, related tools. A particular difficulty of CNMF/CNSC, however, is the high demand on computing power and memory, which can prohibit their application to large scale tasks. In this paper, we propose an online algorithm for CNMF and CNSC, which processes input data piece-by-piece and updates the learned patterns after the processing of each piece by using accumulated sufficient statistics. the online CNSC algorithm remarkably increases converge speed of the CNMF/CNSC pattern learning, thereby enabling its application to large scale tasks.
Shape or color based moment invariants are conventional pattern sensitive features in the object recognition and image description. However, the existing moment invariants are not robust because they handle simplified...
详细信息
暂无评论