Abstract: An underwater rescue system, including a submersible and a rescue cage. The submersible includes a housing, a turntable and a motor. The turntable is rotatably arranged at an outer surface of the housing. The motor is fixed on the housing and connected to the turntable. The rescue cage includes a connecting rod, a first lifebuoy, a second lifebuoy and an air charging device. A lower end of the connecting rod is provided with an electromagnetic chuck attracting the turntable. An upper end of the connecting rod is connected to the first lifebuoy. A bottom of the first lifebuoy is provided with a load-bearing net. A side of the first lifebuoy is connected to the second lifebuoy by a net bag. The air charging device communicates with the first and second lifebuoys and is configured to charge air to the first and second lifebuoys.

Abstract: The present invention disclosed a method for fabrication of fluorescence-Raman dual enhanced modal biometal substrate. The method comprises the steps of grinding surface of the substrate with different types of sandpapers to remove an oxide layer and smooth the surface of the substrate; wherein a roughness of the surface is less than 0.1 ?m; cleaning the grinded substrate in an ultrasonic bath to remove any impurity; placing the specimen on the stage of an ultrashort laser system; processing the specimen at a certain laser processing parameters by a galvanometer; finally, three-dimensional micro-nano structure is fabricated on the specimen. The technique of the present invention is promising for large-scale commercial application because it is simple and economical, while the enhanced Raman and fluorescence signal is stable and high reproducibility.

Abstract: In some embodiments, the disclosed subject matter involves a system and method relating to using an ambient capture device including a fisheye camera and a microphone array to capture audio and video in an environment, for use in an artificial intelligence (Al) application. The device with fisheye camera may provide approximately a 360° audio and video view, at relatively low cost. An embodiment may utilize a speech and vision fusion model component. The speech and vision fusion model may be trained using deep learning to combine features from many different sources, including available sensor data from the capture device. A long short term memory (LSTM) model may inter or identify features such as, but not limited to: audio direction; vision detection and tracking; voice signature; facial signature; gesture recognition; and object identification. The fusion processing may be performed by a cloud server, enabling the capture device to remain less complex.

Abstract: A parking space navigation method is provided. The method includes obtaining and displaying a parking space distribution map of a parking lot by using a mobile terminal; querying current existing vacant parking spaces of the parking lot from a server; marking the current existing vacant parking spaces on the parking space distribution map; determining a target vacant parking space from the marked current existing vacant parking spaces; obtaining, from the server, a first position of a terminal node associated with the mobile terminal when the terminal node is placed on a vehicle; and drawing, on the parking space distribution map, a navigation route from the first position to the target vacant parking space.

Abstract: The present utility model discloses a touch sensor exposed and touch area and writing area integrated electronic whiteboard. A frame is disposed around an electronic whiteboard body. An inner side wall of the frame is provided with an optical reflection strip. Infrared cameras are disposed in upper and lower frames of the frame. The infrared cameras protrude from the optical reflection strip. The two infrared cameras both face a side of the electronic whiteboard body. The two infrared cameras separate the electronic whiteboard body into a touch area and a writing area. A computer host is embedded into the frame in the touch area. A projector is disposed above the touch area, and the projector is connected to the computer host.

Abstract: Representative embodiments disclose mechanisms to separate and recognize multiple audio sources (e.g., picking out individual speakers) in an environment where they overlap and interfere with each other. The architecture uses a microphone array to spatially separate out the audio signals. The spatially filtered signals are then input into a plurality of separators, so each signal is input into a corresponding signal. The separators use neural networks to separate out audio sources. The separators typically produce multiple output signals for the single input signals. A post selection processor then assesses the separator outputs to pick the signals with the highest quality output. These signals can be used in a variety of systems such as speech recognition, meeting transcription and enhancement, hearing aids, music information retrieval, speech enhancement and so forth.

Abstract: The present utility model discloses a non-edge projection and integrated touch electronic whiteboard, comprising: an electronic whiteboard, infrared cameras, a touch mainboard, and a computer. A frame is disposed around the electronic whiteboard. An inner side wall of the frame is provided with an optical reflection strip. The computer separates the electronic whiteboard into a left writing area, a right writing area, and an intermediate touch area. Two infrared cameras are disposed on a same end of an upper frame and a lower frame of the intermediate touch area. The infrared cameras face the intermediate touch area.

Abstract: The present invention discloses a touch electronic whiteboard with automatic recognition for video switching. A computer host is built in an electronic whiteboard body. A HUB integrator is disposed on a frame of the electronic whiteboard body. A projector is disposed above the electronic whiteboard body. A video switching control module is built in the frame of the electronic whiteboard body. The video switching control module comprises a level detection circuit and a control chip that are connected to each other. The level detection circuit is connected to the HUB integrator. The control chip is separately connected to the projector and the computer host.

Abstract: Techniques and architectures may be used to generate and perform a process using weighted finite-state transducers involving generic input search graphs. The process need not pursue theoretical optimality and instead search graphs may be optimized without an a priori optimization step. The process may result in an automatic speech recognition (ASR) decoder that is substantially faster than ASR decoders the include the optimization step.

Abstract: A parking space navigation method is provided. The method includes obtaining and displaying a parking space distribution map of a parking lot by using a mobile terminal; querying current existing vacant parking spaces of the parking lot from a server; marking the current existing vacant parking spaces on the parking space distribution map; determining a target vacant parking space from the marked current existing vacant parking spaces; obtaining, from the server, a first position of a terminal node associated with the mobile terminal when the terminal node is placed on a vehicle; and drawing, on the parking space distribution map, a navigation route from the first position to the target vacant parking space.

Abstract: Techniques and architectures may be used to generate and perform a process using weighted finite-state transducers involving generic input search graphs. The process need not pursue theoretical optimality and instead search graphs may be optimized without an a priori optimization step. The process may result in an automatic speech recognition (ASR) decoder that is substantially faster than ASR decoders the include the optimization step.

Abstract: An electric melting method for forming metal components provides an electric melting head (6) and a base material (2) being connected to the anode and the cathode of a power supply (12). During the forming of the component, the raw metal wire (1) is sent to the base material (2) and the electric melting head (6) to generate electric arc (9) between the raw wire (1) and the base material (2). The electric arc melts a part of the deposited auxiliary material (3) and creates a molten slag pool (8). Electric current generates the resistance heat and the electroslag heat. The raw wire (1) is molten under the high-energy heat resource composed of the electric arc heat, the resistance heat and the electroslag heat, and thereby creating a molten pool (11) on partial surface of the base material (2).

Abstract: An electric melting method for forming a cylinder of a pressure vessel of nuclear power station, in which an electric melting head and a base material are connected to the anode and cathode of a power supply respectively. During the forming of a metal component, the raw metal wire is sent to a surface of the base material by a feeder and the electric melting head to create the electric arc between the raw wire and the base material, wherein the electric arc melts certain of deposited auxiliary material and crates a molten slag pool; a current creates the resistance heat and the electroslag heat; the raw wire is molten under the high-energy heat resource composed of the electric arc heat, the resistance heat and the electroslag heat, and creates a molten pool on partial surface of the base material.

Abstract: ‘Jinhudie’ is a distinctive variety of Broussonetia Plant, which is characterized by its color of leaves.

Abstract: ‘Chongyang’ is a distinctive variety of Chinese redbud plant, which is characterized by its florescence.

Abstract: ‘Jinfeng’ is a distinctive variety of Broussonetia Plant, which is characterized by its color of leaves.

Abstract: ‘Golden Robinia’ is a distinctive variety of Robinia Linn. Plant, which is characterized by that leaves are kept golden in spring, summer and autumn. The Golden Robinia was discovered by the applicant according to: 1) On Jul. 5, 2008, a five-year-old Golden Locust was mutagenized; and on Jul. 22, 2008, a mutant branch, whose leaves were kept golden, was found; then the mutant branch was grafted onto a Robinia rootstock. 2) In the end of August 2008, it was observed that sprouts of the grafted plant were still golden. 3) In spring of 2009, it was observed that leaves of the mother plant and the grafted planted were still golden; furthermore, the leaves were golden and never turned to green in summer and autumn. 4) In the following years, bud and branch grafting methods were used for multiplying, wherein characteristics thereof were kept still.

Abstract: ‘Colorful Manchurian Catalpa’ is a distinctive variety of Catalpa Linn. Plant, which is characterized by that apical leaf edges have irregular rose-red areas, tender leaf edges have irregular yellow areas, and mature leaf edges have irregular white areas. The Colorful Manchurian Catalpa was discovered by the applicant according to: 1) On May 5, 2011, a mutant branch was found on a perennial Manchurian Catalpa at a road side in Long'an District, Anyang City, Henan Province, China. 2) On May 10, 2011, the mutant branch was bud-grafted onto an underyearling Catalpa rootstock with xylem kept, wherein it was observed that a new grafted plate was consistent with the mother plant in characteristics, and no changes happened. 3) From 2012 to 2014, mixed branches from varied generations were used for multiplying; according to observation, no nursery stock mutated, growth was strong, characteristics were kept, and performances were sufficient.

Abstract: ‘Jinfeng’ is a distinctive variety of Broussonetia Plant, which is characterized by its color of leaves.

Abstract: ‘Jinhudie’ is a distinctive variety of Broussonetia Plant, which is characterized by its color of leaves.