Abstract: A lightning protection system of a wind turbine (10) comprises electrically grounded receptors (51, 52) on turbine blades. A corresponding electrical ground connection must be tested regularly. For such a test, an unmanned aerial vehicle or UAV (4) places an electrode (41) near a receptor (51, 52). Using a high voltage generator (42), a high voltage is created between the electrode (41) and the receptor (51, 52), causing an electrical arc (45) between the two. A corresponding current through the lightning protection system (5) indicates whether the ground connection is in order. Thanks to the high voltage used, the electrical arc (45) can bridge a gap between the electrode (41) and receptor (51, 52) of at least several millimetres. It is not necessary to establish a direct physical contact between the electrode (41) and the receptor (51, 52). This simplifies controlling the UAV and speeds up the testing procedure.

Abstract: A method for testing a lightning protection system of a wind turbine comprises the steps of positioning, with an unmanned aerial vehicle (UAV) (4), an electrode (41) near a receptor (51, 52) of a lightning protection system (LPS) (5) of a turbine (10); creating, with a high voltage generator (42), a high voltage between the electrode (41) and the receptor (51, 52); thereby creating an electrical arc (45) between the electrode (41) and the receptor (51, 52); detecting, with a current detection device (46), a current caused by the arc (45) flowing through the lightning protection system (5).

Abstract: A method and a system are provided for searching content (e.g., text, metadata and/or a fingerprint, etc.). In one example, the system receives content and a query for matching the content. The content includes computer readable data. The system generates a feature vector for the content. Generating the feature vector comprises generating a signal from the content, generating a spectrogram from the signal, and generating the feature vector from the spectrogram. The system searches for at least one feature vector that matches the feature vector for the content.

Abstract: An input feature vector is computed from an input text record, the input feature vector comprising one or more features, each feature including a subsequence of characters and a frequency of occurrence of the associated subsequence. A input fingerprint is generated out of the input feature vector by choosing one or more features with non-zero frequencies and alphabetizing the features chosen. One or more input indices are generated by alphabetizing features in the input fingerprint and concatenating features occurring in subsequent locations of the input fingerprint. The input text record is matched against a target text record if (1) one or more of the input indices match a target index corresponding to the target text record and (2) the corresponding input fingerprint matches a target fingerprint corresponding to the target text record. The target text record is outputted as a search result if it matches the input text record.

Abstract: An input feature vector is computed from an input text record, the input feature vector comprising one or more features, each feature including a subsequence of characters and a frequency of occurrence of the associated subsequence. A input fingerprint is generated out of the input feature vector by choosing one or more features with non-zero frequencies and alphabetizing the features chosen. One or more input indices are generated by alphabetizing features in the input fingerprint and concatenating features occurring in subsequent locations of the input fingerprint. The input text record is matched against a target text record if (1) one or more of the input indices match a target index corresponding to the target text record and (2) the corresponding input fingerprint matches a target fingerprint corresponding to the target text record. The target text record is outputted as a search result if it matches the input text record.

Abstract: A system for discovering content and metadata includes a processor communicatively coupled to a communication network and a database. The processor determines whether an end portion of a portion of content has been received based on the portion of content and/or metadata. The processor generates a content fingerprint based on the portion of content if the end portion has been received. The content fingerprint and/or the metadata are stored in the database.

Abstract: A system includes a processor that receives, via a communication channel, a portion of content associated with time-localized metadata. The time-localized metadata and a tag mode identifier are retrieved from a database. A tag mode associated with the portion of content is determined based on the time-localized metadata and/or the tag mode identifier. The processor implements a feature based on the time-localized metadata and the tag mode.

Abstract: A system includes a processor that retrieves, from a first database, a portion of content associated with time-localized metadata. The time-localized metadata and a tag mode identifier are retrieved from a second database. A tag mode associated with the portion of content is determined based on the time-localized metadata and/or the tag mode identifier. The portion of content and/or the time-localized metadata are transmitted to a user device based on the tag mode. In some embodiments, a reconstruction mode associated with the portion of content is determined based on the time-localized metadata and/or a reconstruction mode identifier. The portion of content, the time-localized metadata, attribute information, and/or mapping information are transmitted to the user device based on the reconstruction mode.

Abstract: Providing a recommendation of a media item. Media item identifiers and corresponding nonzero media item attribute strengths for a target media item attribute are retrieved from a database. Each of the retrieved media item attribute strengths are randomized, resulting in randomized media item attribute strengths. One of the media item identifiers is selected according to predetermined selection criteria. A recommendation of a media item corresponding to the selected media item identifier is transmitted to a user device over a network.

Abstract: A processor retrieves, from a database, an identifier of a first media item rated by a target user, user identifiers of users having provided media item ratings for the first media item, and the media item ratings provided by the users, correspondingly. Each of the media item ratings is randomized, resulting in randomized media item ratings. A user identifier corresponding to a maximum of the randomized media item ratings is selected from the user identifiers. Additional media item ratings associated with the selected user identifier are retrieved from the database. Each of the additional media item ratings are randomized, resulting in additional randomized media item ratings. An identifier of a recommended media item corresponding to a maximum of the randomized media item ratings and the additional randomized media item ratings is selected. Metadata associated with the recommended media item is transmitted to a user device over a network.

Abstract: A method and a system are provided for recommending a media item (e.g., movie and/or television program). In one example, the system receives a seed media item (e.g., seed movie and/or seed television program). A seed media item is an audio-visual recording that is identified as being of interest to a user. The system recognizes one or more recognized audio tracks (e.g., audio tracks and/or songs) from the seed media item. A recognized audio track is an audio recording that is contained in the seed media item. The system retrieves audio track metadata associated with at least one of the recognized audio tracks from the seed media item. Audio track metadata are data that describes an audio recording. The system recommends one or more recommended media items based on the audio track metadata. A recommended media item is an audio-visual recording (e.g., movie and/or television program).

Abstract: An exemplary geometric image compression method includes identifying an image having a plurality of pixels, and iteratively obtaining largest geometric shapes by using corresponding reference points until all pixels of the image are processed. Each geometric shape includes at least a substantial portion of pixels with color values corresponding to a color value of a corresponding reference point. The method further includes generating, for each obtained geometric shape, identifying information including data indicative of a color value of the corresponding reference point or the average of the color values of at least the substantial portion of pixels within the geometric shape, data indicative of a location of the geometric shape, and data indicative of a size of the geometric shape, and storing compressed image data comprising identifying information of each obtained geometric shape.

Abstract: A processor retrieves, from a database, an identifier of a first media item rated by a target user, user identifiers of users having provided media item ratings for the first media item, and the media item ratings provided by the users, correspondingly. Each of the media item ratings is randomized, resulting in randomized media item ratings. A user identifier corresponding to a maximum of the randomized media item ratings is selected from the user identifiers. Additional media item ratings associated with the selected user identifier are retrieved from the database. Each of the additional media item ratings are randomized, resulting in additional randomized media item ratings. An identifier of a recommended media item corresponding to a maximum of the randomized media item ratings and the additional randomized media item ratings is selected. Metadata associated with the recommended media item is transmitted to a user device over a network.

Abstract: A method and a system are provided for recommending a media item (e.g., movie and/or television program). In one example, the system receives a seed media item (e.g., seed movie and/or seed television program). A seed media item is an audio-visual recording that is identified as being of interest to a user. The system recognizes one or more recognized audio tracks (e.g., audio tracks and/or songs) from the seed media item. A recognized audio track is an audio recording that is contained in the seed media item. The system retrieves audio track metadata associated with at least one of the recognized audio tracks from the seed media item. Audio track metadata are data that describes an audio recording. The system recommends one or more recommended media items based on the audio track metadata. A recommended media item is an audio-visual recording (e.g., movie and/or television program).

Abstract: A method and a system are provided for searching content (e.g., text, metadata and/or a fingerprint, etc.). In one example, the system receives content and a query for matching the content. The content includes computer readable data. The system generates a feature vector for the content. Generating the feature vector comprises generating a signal from the content, generating a spectrogram from the signal, and generating the feature vector from the spectrogram. The system searches for at least one feature vector that matches the feature vector for the content.

Abstract: A method and a system are provided for representing content (e.g., text, metadata and/or a fingerprint, etc.). In one example, the system receives content. The content includes computer readable data. The system generates normalized content by normalizing the content. Normalizing the content is a process by which content is transformed to make the content more consistent for computer reading purposes. The system generates content blocks (e.g., words, etc.) from the normalized content. The system generates values for the content blocks by generating a value for each content block. The system generates a signal for the content based on the values of the content blocks. The signal includes a representation of the values versus time. The system generates a spectrogram from the signal, generates a feature vector from the spectrogram, and stores the feature vector in a database.