484 results about "Scene analysis" patented technology

A video communication system and method for operating a video communication system are provided. The video communication system has a video communication device, having an image display device and at least one image capture device, wherein the at least one image capture device acquires video images of a local environment and an individual therein, according to defined video capture settings, an audio system having an audio emission device and an audio capture device; and a computer operable to interact with a contextual interface, a privacy interface, an image processor, and a communication controller to enable a communication event including at least one video scene in which outgoing video images are sent to a remote site. Wherein the contextual interface includes scene analysis algorithms for identifying potential scene transitions and capture management algorithms for providing changes in video capture settings appropriate to any identified scene transitions; and wherein the privacy interface provides privacy settings to control the capture, transmission, display, or recording of video image content from the local environment.

The present invention provides a deep and reinforcement learning-based real-time online path planning method. According to the method, the high-level semantic information of an image is obtained through using a deep learning method, the path planning of the end-to-end real-time scenes of an environment can be completed through using a reinforcement learning method. In a training process, image information collected in the environment is brought into a scene analysis network as a current state, so that an analytical result can be obtained; the analytical result is inputted into a designed deep cyclic neural network; and the decision-making action of each step of an intelligent body in a specific scene can be obtained through training, so that an optimal complete path can be obtained. In an actual application process, image information collected by a camera is inputted into a trained deep and reinforcement learning network, so that the direction information of the walking of the intelligent body can be obtained. With the method of the invention, obtained image information can be utilized to the greatest extent under a premise that the robustness of the method is ensured and the method slightly depends on the environment, and real-time scene walking information path planning can be realized.

Video communication systems and methods for operating the same are provided with the method determining privacy settings for a video communication event; establishing a video communication link with a video communication system at a remote site; capturing video images of a field of view of at least one image capture device; determining a video context based at least in part upon a determined contextual classification for the communication event; analyzing the captured video images to identify privacy sensitive areas in the captured video images, using a scene analysis algorithm which examines a content of the video images and image context characteristics of the video images based upon the determined privacy settings; modifying the image characteristics in identified privacy sensitive image areas of captured video images, to control the image and image context characteristics to an extent consistent with the determined privacy settings; and transmitting the modified video images.

Video communication devices and methods are provided. The device has an image display device; a first image capture device that acquires video images depicting a wide field of view within the local environment; a second image capture device that acquires video images depicting a narrow field of view within the wide field of view; a communication system that is adapted to use a communication network to transmit outgoing video images; and a computer causing the capture of video images during a communication event, and further being adapted to identify a video context for the communication event; wherein said computer uses a scene analysis algorithm to examine concurrent video images from the image capture device based upon the identified video context to determine the extent to which video images from the first image capture device and the second image capture device are to be incorporated into the outgoing video images.

In a method of cognitive communication for non-interfering transmission, wherein the improvement comprises the step of conducting radio scene analysis to find not just the spectrum holes or White spaces; but also to use the signal classification, machine learning and prediction information to learn more things about the existing signals and its underlying protocols, to find the Gray space, hence utilizing the signal space, consisting of space, time, frequency (spectrum), code and location more efficiently.In a method of cognitive jamming where smart and energy efficient jamming techniques are suggested based on sensing, classification and machine learning of the existing signals.

Methods, systems, and apparatuses are described for improved multi-microphone source tracking and noise suppression. In multi-microphone devices and systems, frequency domain acoustic echo cancellation is performed on each microphone input, and microphone levels and sensitivity are normalized. Methods, systems, and apparatuses are also described for improved acoustic scene analysis and source tracking using steered null error transforms, on-line adaptive acoustic scene modeling, and speaker-dependent information. Switched super-directive beamforming reinforces desired audio sources and closed-form blocking matrices suppress desired audio sources based on spatial information derived from microphone pairings. Underlying statistics are tracked and used to updated filters and models. Automatic detection of single-user and multi-user scenarios, and single-channel suppression using spatial information, non-spatial information, and residual echo are also described.

In a method of cognitive communication a system for generating non-interfering transmission, includes conducting radio scene analysis to find grey space using external signal parameters for incoming signal analysis without having to decode incoming signals.

Video communication systems and methods for operating the same are provided. The system has a computer cooperating with at least one image capture device which acquires video images of a local environment having an individual therein, an audio system, an image processor and a communication controller to acquire video images based upon the acquired video images and to cause output video images to a remote individual during the communication event; said computer further providing a contextual interface including an intra-scene analysis algorithm for identifying potential intra-scene transitions during the communication event and a scene capture management algorithm for determining intra-scene adjustments in video capture settings when an intra-scene transition is detected; and the contextual interface further including a transition test with an inter-scene analysis algorithm, for identifying potential inter-scene transitions in the activities of the individuals, and a transition process structure for determining inter-scene adjustments in video capture settings.

An electronic camera is provided with an image-capturing device for photographing (73, 26) that captures an image of a subject image passed through a taking lens and outputs image data, an image-capturing device for scene analysis (86) that is provided at a position conjugate with the position of the image-capturing device for photographing relative to the taking lens and outputs scene analysis image data by receiving the light of the subject image and a scene detection processing circuit (35) that determines a white balance adjustment gains based upon the scene analysis image data output by the analytical image-capturing device. Using the white balance adjustment gains determined by the scene detection processing circuit, white balance adjustment is performed on the image data output by the image-capturing device for photographing.

The invention discloses a real-time security early warning method based on complex event processing. The method specifically comprises the following steps: step (1), using a formalized engine to perform log field segmentation on collected security data, performing normalization on fields according to field requirements, and associating knowledge base information according to the fields that are expected to be outputted; step (2), using a data flow semantic analysis engine to perform data context analysis according to a complex event instance to be modeled as a scene, and analyzing mapping stream data based on a standardized analysis field template; and step (3), using a security analysis model calculation engine, in an analysis rule calculation module, analyzing according to scenes based on a point event, an edge event and an interval event, and generating an early warning event. The real-time security early warning method based on the complex event processing provided by the invention realizes the multi-angle association analysis of original log data through a configurable normalization rule, a semantic recognition rule and a security analysis rule, and discovers unknown threats and carries out early warning timely.

Video communication systems and methods for operating the same are provided. The system has a computer cooperating with at least one image capture device which acquires video images of a local environment having an individual therein, an audio system, an image processor and a communication controller to acquire video images based upon the acquired video images and to cause output video images to a remote individual during the communication event; said computer further providing a contextual interface including an intra-scene analysis algorithm for identifying potential intra-scene transitions during the communication event and a scene capture management algorithm for determining intra-scene adjustments in video capture settings when an intra-scene transition is detected; and the contextual interface further including a transition test with an inter-scene analysis algorithm, for identifying potential inter-scene transitions in the activities of the individuals, and a transition process structure for determining inter-scene adjustments in video capture settings.

The invention discloses an online learning system which comprises a courseware database module, a management module, and a student learning situation monitoring module. The student learning situation monitoring module comprises a face recognition module, a facial expression analysis module, a video scene analysis module, a sound recognition module, a monitoring video storage module, and a video acquisition module, and the like. The learning situation monitoring module can be applied to monitor the student learning and exam situations in real time, thereby preventing other people from taking learning and tests for someone, and improving the learning efficiency of students.

In order to safeguard a hazardous area, in particular the hazardous area of an automatically operating machine, images of the hazardous area are recorded with at least a first and a second image recording unit that are arranged offset from one another. The images of the scene are subjected to three-dimensional scene analyses using different three-dimensional scene analysis methods, and the hazardous area is safeguarded when at least one of the scene analysis methods detects a foreign object.

A video communication system and method for operating a video communication system are provided. The video communication system has a video communication device, having an image display device and at least one image capture device, wherein the at least one image capture device acquires video images of a local environment and an individual therein, according to defined video capture settings, an audio system having an audio emission device and an audio capture device; and a computer operable to interact with a contextual interface, a privacy interface, an image processor, and a communication controller to enable a communication event including at least one video scene in which outgoing video images are sent to a remote site. Wherein the contextual interface includes scene analysis algorithms for identifying potential scene transitions and capture management algorithms for providing changes in video capture settings appropriate to any identified scene transitions; and wherein the privacy interface provides privacy settings to control the capture, transmission, display, or recording of video image content from the local environment.

The invention relates to a dynamic WEB 3D virtual reality scene construction method and system based on model componentization. The system comprises a model component server, a scene editing server, a scene dynamic loading server and a client, wherein models are standardized according to unified standards, after the models are componentized, a component library is formed, and a unified material library is provided for posterior development of 3D virtual reality items; the scene dynamic loading server is used for optimizing display of Web 3D scenes and transmitting dynamic scene model data to a user to be loaded on batches; by receiving the scene analysis result of the scene dynamic loading server, model resources are downloaded by the client in multiple queues according to dynamic loading requirements, and the scenes are rendered according to the priority of scene models and the depending relation between the scene models. After the models are componentized, the models can be multiplexed in the virtual reality items, using redundancy of the models in scene construction is reduced, the 3D engine efficiency and the network transmission efficiency are both promoted, hardware cost for a user to experience 3D virtual reality is reduced, and user experience is improved.

The invention discloses a video abstraction generating method based on sketches, which belongs to the field of man-machine interaction. The method comprises the following steps: (1) carrying out shot segmentation and shot screening on the video to obtain key frames; (2) carrying out character analysis and scene analysis on the key frames, and determining the semantic significance level of each key frame according to the character analysis; (3) calculating an image quality assessed value of each key frame, and screening the key frames according to the image quality assessed value and the semantic significance level value; (4) generating a video object sketch according to each key frame determined in step (3); (5) resetting the size of the corresponding sketch according to the size of the abstraction drawing region and the significance level of the video contents corresponding to each sketch; and (6) determining the position of each sketch by utilizing the scene analysis result, generating connecting lines among the sketches, and connecting the sketches to generate a video abstraction. By using the method of the invention, the sketch abstraction can intensively and effectively reflect the main plot semanteme of the video, thereby conforming to the perception habits of users.

Estimating the dynamic states of a real-world object over time using a camera, two dimensional (2D) image information and a combination of different measurements of the distance between the object and the camera. The 2D image information is used to track a 2D position of the object as well as its 2D size of the appearance and change in the 2D size of the appearance of the object. In addition, the distance between the object and the camera is obtained from one or more direct depth measurements. The 2D position, the 2D size, and the depth of the object are coupled to obtain an improved estimation of three dimensional (3D) position and 3D velocity of the object. The object tracking apparatus uses the improved estimation to track real-world objects. The object tracking apparatus may be used on a moving platform such as a robot or a car with mounted cameras for a dynamic visual scene analysis.

According to an embodiment of the present invention, security exposure analysis of wireless network within a selected local geographic area is provided. A computer model of the selected local geographic region comprising a layout is generated. Information regarding wireless network components is provided to the computer model. Using the computer model, signal intensity characteristics of at least one of the wireless network components are determined over at least a portion of the selected geographic region. Based at least on the signal intensity characteristics, security exposure information associated with the wireless network is determined. The security exposure information is graphically displayed on the computer screen in relation to the layout of the selected geographic region. The security exposure information includes sniffer detection and prevention coverage, access point vulnerability regions, and signal uncertainty and variability views.

A pair of spectacles that can automatically change its power so that a fixation region of interest (ROI) of the user is always in focus. The automatic accommodative spectacle device includes focusing elements, scene sensors, scene analyzer, focus engine, focusing element controller, and power supply. The scene analyzer determines the fixation ROI in the scene by analyzing the scene images. The fixation ROI is used to determine powers for the focusing elements in order to bring the fixation ROI into focus. The focusing element controller carries out the needed optical power adjustment to apply to the focusing elements. Optional light sources may be provided to the scene sensors. Additional optional eye sensor(s), eye light source and line of sight detector can be used to help the scene analyzer extract the fixation ROI.

The invention discloses a system and a method for unified configuration of network equipment, and mainly solves the problems of huge workload and the lack of security function for configuring the network equipment in the prior configuration method. A self-designed command set is used for the unified configuration of network equipment from different factories and in different models. The system mainly comprises a server, an agent and the network equipment, wherein the server is loaded with a command event acquisition plug-in, a reply event acquisition plug-in, an event format extension plug-in, a scene analysis plug-in, an event response plug-in, an equipment information file, a command conversion information file and a command conversion rule file. The elf-designed command is converted into a special command which can be executed in the target network equipment by the server, and forwarded to the network equipment for execution through the agent, and after execution, the output information of the network equipment is transmitted to the agent for processing and then returned to the server. The system and the method have the advantages of convenient centralized and unified configuration of the network equipment, and can be widely applied in the field of computer network management.

The invention discloses an intelligent air conditioner controller, a control method and an air conditioner. The air conditioner controller comprises an image input unit used for receiving an input activity image of people in the environment where the air conditioner is located, a scene mode bank used for correspondingly storing the category of the activity of the people and scene modes so as to be formed, a scene analyzing unit used for identifying the category of the activity of the people according to the input activity image of the people and determining the scene mode which the activity image belongs to, and a control unit used for controlling operation of the intelligent air conditioner according to the determined scene mode. By means of the intelligent air conditioner controller, the control method and the air conditioner of the embodiment of the invention, the intelligent automatic mode switching can be provided, the current scene can be detected and judged in real time, and the operation mode of the air conditioner can be automatically regulated.

A method for is provided for creating a shadow-reduced image from a captured image for distinguishing a clear path of travel. Each pixel of a captured input image is plotted according to a two dimensional logarithmic graph. A specific color set relating to an associated color value of a clear path. A linear illumination-invariant axis is determined as a function of the specific color set. An illumination direction for the linear illumination-invariant axis is determined. A log-chromaticity value of each plotted pixel of the specific color set is projected on the axis. Edges in the input image and the illumination-invariant image domain are identified. The identified edges of the input image are compared to identify edges in the illumination-invariant image domain. A determination is made whether a shadow edge is present in response to comparing the edges. A shadow-reduced image is generated for scene analysis by a vehicle vision-based system.

The invention provides a voice enhancement system. A specific double microphone structure is used to simulate hearing scene analytical capability of human ears so as to realize voice enhancement. And the voice enhancement system has a characteristic of being independent of a noise type and can be widely applied for voice enhancement in various noise environments. A principle of calculation on a hearing scene analysis is utilized to enable target voice to be separated from background noises, so that denoising is realized; and therefore, it can be realized that the voice enhancement is independent of a concrete type of the noise, the number of various noise sources, a topological structure of space between a target sound source and a noise source; and moreover, a good denoising effect is realized and the good voice quality can be ensured.

According to an air conditioner control method based on indoor scene analysis, an image collecting device is arranged on an air conditioner so that an indoor image can be collected; comprehensive analysis of the human body posture, the human body motion trajectory and the position relationship between the human body and furniture is conducted on the collected indoor image through an artificial intelligent network so that a sleep scene, a dining scene, a party scene, a motion scene and a leisure scene in a room can be identified; and the temperature and the air blowing mode of the air conditioner are automatically controlled according to the identified different scenes.

The invention relates to the technical field of electronic information, and particularly relates to a real-time interactive online learning platform comprising a courseware database module and a management module, etc. The real-time interactive online learning platform also comprises a student learning situation monitoring module. The student learning situation monitoring module comprises a human face identification module, a facial expression analysis module, a video scene analysis module, a sound identification module, a monitoring video storage module and a video acquisition module, etc. The real-time interactive online learning platform is reasonable in structural design. The learning and testing situations of students can be monitored in real time by using the learning situation monitoring module so that the situations of learning replacement and testing replacement can be voided, and the learning effect of the students is enhanced.

The invention relates to a scene analysis method and voltage regulation policy-based distributed photovoltaic optimization planning method. The method comprises the steps of 1, performing scene analysis by utilizing a scene analysis method, representing real running states of a distributed power supply and a load with high-dimensional vectors, and performing scene compression through a clustering method; and 2, establishing a dual-layer optimization planning method for the distributed power supply of a power distribution network: (1) in upper-layer planning, constructing multiple targets by minimization of network loss of the power distribution network, maximization of earnings of a distributed power operator and the like, and planning access capacity of the distributed power supply; and (2) adopting different scheduling policies in a lower-layer scheduling model, comparing non adoption of a voltage regulation policy, adoption of the voltage regulation policy and a voltage regulation policy with reactive compensation, and planning the distributed power supply.