216 results about "Separation result" patented technology

LED packages, and LED lamps and bulbs, are disclosed that are arranged to minimize the CRI and efficiency losses resulting from the overlap of conversion material emission and excitation spectrum. In different devices having conversion materials with this overlap, the present invention arranges the conversion materials to reduce the likelihood that re-emitted light from a first conversion materials will encounter the second conversion material to minimize the risk of re-absorption. In some embodiments this risk is minimized by different arrangements where there is separation between the two phosphors. In some embodiments this separation results less than 50% of re-emitted light from the one phosphor passing into the phosphor where it risks re-absorption.

The invention discloses a MR (Measurement Report) data indoor and outdoor separation method based on a statistic model. The MR data indoor and outdoor separation method comprises the steps of: aiming at MR sampled data of each cell of an outdoor macro station, carrying out characteristic value statistics of received signals, which comprises statistics of a main cell level; and carrying out separation of a mixed Gaussian distribution and probability calculation, and according to an indoor probability, obtaining a corresponding indoor and outdoor separation result. Moreover, the MR data indoor and outdoor separation method supports combination of various factors of the main cell level, a main and adjacent level difference, the number of adjacent cells and the like, and by a combined model, an indoor and outdoor judgment result is judged and output; and judgment methods at other angles also can be integrated. According to the MR data indoor and outdoor separation method disclosed by the invention, data characteristics which are shown by signal intensity and a signal number of the MR data in a statistical sense and are combined by two indoor and outdoor normal distributions are utilized, and probabilities that different signal intensity and different signal numbers occur in a door or outside a door can be calculated, so that the method of separating the MR data according to signal characteristics is more feasible, and accuracy of indoor and outdoor separation of the MR data is improved.

A voice recognition system (10) for improving the toughness of voice recognition for a voice input for which a deteriorated feature amount cannot be completely identified. The system comprises at least two sound detecting means (16a, 16b) for detecting a sound signal, a sound source localizing unit (21) for determining the direction of a sound source based on the sound signal, a sound source separating unit (23) for separating a sound by the sound source from the sound signal based on the sound source direction, a mask producing unit (25) for producing a mask value according to the reliability of the separation results, a feature extracting unit (27) for extracting the feature amount of the sound signal, and a voice recognizing unit (29) for applying the mask to the feature amount to recognize a voice from the sound signal.

The invention provides a method for separating a vehicle running track, and includes the steps of: (1) acquiring satellite positioning data of operation of a vehicle in a set period of time, the satellite positioning data including the acquisition time, the vehicle position and the passenger carrying state corresponding to the satellite positioning data; (2) screening the acquired satellite positioning data according to the passenger carrying sate of each satellite positioning data and forming a corresponding passenger carrying running track according to a screening result; and (3) determining the positions of time jump points and space jump points of the passenger carrying running track, and dividing the passenger carrying track into a plurality of sub-running tracks according to the positions of the time jump points and the space jump points. The method for separating the vehicle running track completes division in sequence according to the time jump points and the space jump points, is easy to realize, and fully considers the characteristic of the vehicle running track, thereby enabling a finally obtained track separation result to be more accurate, and accuracy of a result of performing motion mode mining using the track separation result can be effectively improved.

The invention discloses an X-ray chest radiograph image quality determination method and device. The method comprises: conducting normalization processing on an X-ray chest radiograph image; inputtingthe normalized image into a deep learning model for image separation and recognition to obtain at least one separation result; wherein the deep learning model at least comprises one of the followingformulas: a pulmonary lobe segmentation model, a spine segmentation model, a shoulder blade segmentation model and a foreign matter detection model; and determining the image quality of the X-ray chest radiograph based on the separation result. The X-ray chest radiograph image quality is fully automatically evaluated, and the image quality determination speed is high. Moreover, a technician can behelped to control the image quality of the chest radiograph, and the radiograph reading accuracy is indirectly improved.