Abstract: The present invention provides a method and an apparatus for detecting a salient region of an image. Classification processing is performed on a test image according to an image feature vector of the test image by using a classifier obtained by means of pre-training, so as to obtain a classification label, where the classification label is used to indicate a salience detection algorithm for detecting a salient region of the test image. Salience detection is performed on the test image by using the salience detection algorithm indicated by the classification label, so as to obtain the salient region of the test image. Because a salience detection algorithm with the best detection effect is acquired by using the image feature vector of the test image, to detect the salient region of the test image, accuracy of salience detection is improved.

Abstract: Embodiments of the present invention provide a video synopsis method and apparatus, and relate to the field of video synopsis. The method includes: extracting moving objects and moving tracks of the moving objects from an original video; calculating a spatio-temporal redundancy between the extracted moving tracks, and determining a synopsis ratio of the moving tracks according to the spatio-temporal redundancy; and changing rates of the moving tracks according to the synopsis ratio of the moving tracks. In the present invention, the spatio-temporal redundancy between the moving tracks is analyzed, and synopsis and rate changing are performed on corresponding tracks according to the spatio-temporal redundancy obtained by means of analysis to obtain different representations of different tracks, thereby improving user experience.

Abstract: A method includes collecting vehicle health data from a plurality of vehicles. A peer group is identified among the plurality of vehicles. The collected vehicle health data from the peer group into a collaborative vehicle health model, the collaborative vehicle health model being applicable to a current vehicle to predict a state of at least a component of the current vehicle.

Abstract: An embodiment contemplates a method of determining a state-of-charge of a battery for a vehicle. (a) An OCV is measured for a current vehicle ignition startup after ignition off for at least eight hours. (b) An SOCOCV is determined for the current vehicle ignition startup. (c) An SOCOCV—est is is determined for a current vehicle ignition startup. (d) A determination is made whether the difference in the SOCOCV for the current startup and the SOCOCVest for the current startup is less than a predefined error bound using. Steps (a)-(d) is performed in response to the difference being greater than the predefined error; otherwise, determining an ignition-off current for the current vehicle ignition startup as a function of the SOCOCV of the current vehicle ignition startup and previous vehicle ignition startup, and a SOC based on current integration over time. Determining an SOCest of the current vehicle ignition startup using the processor.

Abstract: A method includes collecting state of health (SOH) data and usage data from a plurality of vehicles. A peer group is identified among the vehicles. A vehicle health prognosis is generated for each vehicle of the peer based on the collected SOH and usage data. The vehicles of the peer group are ranked based on the generated vehicle health prognosis and the rank is reported to an output device that is associated with each vehicle or with a user of each vehicle. If evaluation of the ranking indicates that the health prognosis of a vehicle of the peer group is improvable by modifying vehicle usage, an alert is issued to a user of that vehicle.

Abstract: A method includes collecting vehicle health data from a plurality of vehicles. A peer group is identified among the plurality of vehicles. The collected vehicle health data from the peer group into a collaborative vehicle health model, the collaborative vehicle health model being applicable to a current vehicle to predict a state of at least a component of the current vehicle.

Abstract: A method includes collecting state of health (SOH) data and usage data from a plurality of vehicles. A peer group is identified among the vehicles. A vehicle health prognosis is generated for each vehicle of the peer based on the collected SOH and usage data. The vehicles of the peer group are ranked based on the generated vehicle health prognosis and the rank is reported to an output device that is associated with each vehicle or with a user of each vehicle. If evaluation of the ranking indicates that the health prognosis of a vehicle of the peer group is improvable by modifying vehicle usage, an alert is issued to a user of that vehicle.