Abstract: The present application relates to a lexicon learning-based heliumspeech unscrambling method in saturation diving. In a system including divers, a correction network, and an unscrambling network, a common working language lexicon for saturation diving operation is established and is read by the divers respectively in different environments, to generate supervision signals and vector signals of the correction network, and the correction network learns heliumspeeches of the different divers at different diving depths to obtain a correction network parameter, and corrects a heliumspeech of a diver to obtain a corrected speech; and the unscrambling network learns the corrected speech and completes unscrambling of the heliumspeech.

Abstract: A labeling method is provided. In the method, target image data of a target image is acquired from an offline storage device. The target image data is encrypted based on a first target account. A permission to load the target image data is verified based on a second target account logged in to a labeling client. The target image is loaded in the labeling client in response to the permission to load the target image being verified when the second target account logged in to the labeling client matches the first target account. A label for the target image is received. Further, the label for the target image is transmitted to a server. The server is configured to store the label in association with the target image. Apparatus and non-transitory computer-readable storage medium counterpart embodiments are also contemplated.

Abstract: The present application discloses a method and a system for saturation diving heliumspeech unscrambling based on multi-objective optimization. In a system including a diver and a filter at least, a working language phonetic symbol library and a common working word library for divers are constructed. The divers read them one by one, and a phonetic symbol standard speech library, a phonetic symbol heliumspeech library and a common working word speech library are generated. The filter uses the multi-objective optimization algorithm to design its impulse response coefficients, corrects and unscrambles the tagged and sampled heliumspeech signal word by word, and continuously updates the impulse response coefficients to complete the perfect heliumspeech unscrambling.

Abstract: The present application discloses a method and a system for saturation diving heliumspeech unscrambling based on multi-objective optimization. In a system including a diver and a filter at least, a working language phonetic symbol library and a common working word library for divers are constructed. The divers read them one by one, and a phonetic symbol standard speech library, a phonetic symbol heliumspeech library and a common working word speech library are generated. The filter uses the multi-objective optimization algorithm to design its impulse response coefficients, corrects and unscrambles the tagged and sampled heliumspeech signal word by word, and continuously updates the impulse response coefficients to complete the perfect heliumspeech unscrambling.

Abstract: The present application relates to a lexicon learning-based heliumspeech unscrambling method in saturation diving. In a system including divers, a correction network, and an unscrambling network, a common working language lexicon for saturation diving operation is established and is read by the divers respectively in different environments, to generate supervision signals and vector signals of the correction network, and the correction network learns heliumspeeches of the different divers at different diving depths to obtain a correction network parameter, and corrects a heliumspeech of a diver to obtain a corrected speech; and the unscrambling network learns the corrected speech and completes unscrambling of the heliumspeech.

Abstract: A data processing method and apparatus are provided. The method includes: predicting, based on an initial image recognition model, initial aided annotation results of original images, and acquiring initial standard annotation results; adjusting, based on a first initial standard annotation result and a first initial aided annotation result, model parameters in the initial image recognition model so as to generate an updated image recognition model; predicting, based on the updated image recognition model, an updated aided annotation result of a second original image, and acquiring an updated standard annotation result; and determining, in response to determining that the updated image recognition model satisfies a model convergence condition based on the updated aided annotation result and the updated standard annotation result, the updated image recognition model as a target image recognition model.

Abstract: A broadband super-Rayleigh speckle correlated imaging spectral camera based on dispersion compensation is provided. The imaging scheme comprises, but is not limited to, a pre-compensation scheme, a post-compensation scheme, or a pre-post joint compensation scheme. The device comprises components such as a pre-imaging module, a light filter, a phase modulation module, a relay imaging module, an area array detector, and a computer. According to the present invention, the super-Rayleigh speckle modulation in a broadband is realized by matchining the dispersion characteristic of the pre-imaging module or the relay imaging module with the phase modulation module, which is applied to the correlated imaging spectral camera, so that the imaging quality of the correlated imaging spectral camera at a low signal-to-noise ratio is improved.

Abstract: A labeling method is provided. In the method, target image data of a target image is acquired from an offline storage device. The target image data is encrypted based on a first target account. A permission to load the target image data is verified based on a second target account logged in to a labeling client. The target image is loaded in the labeling client in response to the permission to load the target image being verified when the second target account logged in to the labeling client matches the first target account. A label for the target image is received. Further, the label for the target image is transmitted to a server. The server is configured to store the label in association with the target image. Apparatus and non-transitory computer-readable storage medium counterpart embodiments are also contemplated.

Abstract: The disclosure discloses microbial conductive ceramics and a preparation method and application thereof, and belongs to the technical field of microorganisms and the technical field of semiconductor materials. The disclosure is based on ordinary insulating macroporous ceramics, using the means of cell immobilization and the principle of microbial adsorption, to prepare the microbial conductive ceramics including macroporous ceramics, microbes immobilized on the macroporous ceramics and metal ions adsorbed to the microbes. The microbial conductive ceramics have excellent performance, and the conductivity of the microbial conductive ceramics can reach 2.91×106 S/m. At the same time, the cost of the microbial conductive ceramics is low, only 10% of the cost of conductive ceramics with the same conductivity.

Abstract: A random grating based compressive sensing wideband hyperspectral imaging system comprising a front imaging component, an optical splitting component, more than two branches, and a computer, with each branch having an exit pupil transformation component and a compressive spectral imaging component. The system is based on compressive sensing theory, makes use of the correlation between spectra, increases compression rate on the spectral dimension, realizes three-dimensional compressive collection of wideband spectral image data, and greatly reduces data collection amount. The system is capable of measuring spectral image information on a wideband spectrum with a single exposure, and may obtain high spatial resolution and high spectral resolution by means of designing the random gratings on the various spitting paths.

Abstract: A flicker detection apparatus and method where flickers in the video images are detected based on the mean detection values obtained from pairs of neighboring frames in the current frame and the former n frames, which is able to perform flicker detection efficiently, and the detection results are accurate.

Abstract: A random grating based compressive sensing wideband hyperspectral imaging system comprising a front imaging component, an optical splitting component, more than two branches, and a computer, with each branch having an exit pupil transformation component and a compressive spectral imaging component. The system is based on compressive sensing theory, makes use of the correlation between spectra, increases compression rate on the spectral dimension, realizes three-dimensional compressive collection of wideband spectral image data, and greatly reduces data collection amount. The system is capable of measuring spectral image information on a wideband spectrum with a single exposure, and may obtain high spatial resolution and high spectral resolution by means of designing the random gratings on the various spitting paths.

Abstract: A method and apparatus for detecting a traffic state and electronic equipment where the apparatus includes: a first calculating unit configured to detect a contour of an object from a region of interest of a traffic surveillance image to generate a contour image, and calculate an occupation ratio of the contour in the region of interest based on the contour image; a first detecting unit configured to detect whether there exists an object of a predetermined size moving in the region of interest, and determine a motion state of a vehicle in the region of interest; and a first determining unit configured to determine a traffic state according to the occupation ratio and the motion state. The embodiments of the present disclosure determine a traffic state according to the number of vehicles and motion states of the vehicles in the traffic surveillance image.

Abstract: A flicker detection apparatus and method where flickers in the video images are detected based on the mean detection values obtained from pairs of neighboring frames in the current frame and the former n frames, which is able to perform flicker detection efficiently, and the detection results are accurate.

Abstract: An apparatus and method for controlling traffic lights of a crossing where the traffic lights are controlled based on the jam states of the roads and according to jam levels of the roads, weights of the roads and the number of vehicles passing the crossing, which are fast in response and accurate in estimation, and the traffic lights can be quickly and efficiently controlled, thereby efficiently alleviating or solving the problem of traffic jam.