3 results about "Data sanitization" patented technology

The embodiment of the application belongs to the field of artificial intelligence, and relates to a title generation method based on artificial intelligence, comprising the following steps: acquiring a pre-acquired title data set; cleaning titles in the title data set based on multiple data cleaning rules to obtain target text data; using the target text data as training data to train a pre-trained language model based on an adversarial training strategy and a noise reduction strategy, to obtain a title generation model; acquiring a target text to be recognized; inputting the target text into the title generation model, and processing the target text by the title generation model to generate a corresponding target title. The application also provides a title generation device based on artificial intelligence, a computer device and a storage medium. In addition, the application also relates to blockchain technology, and the target title can be stored in the blockchain. The application uses the title generation model to quickly and accurately generate the target title corresponding to the target text to be recognized, and ensures the accuracy of the generated target title.

The application discloses a data cleaning method and device, equipment and a storage medium. The method comprises the following steps: determining the logical blocks corresponding to each partition of the storage medium to be cleaned and the corresponding target cleaning strategy; performing hierarchical cleaning on the logical blocks based on the target cleaning strategy of each logical block until the storage medium to be cleaned is completely cleaned; wherein the target cleaning strategy is one of a first strategy for full cleaning, a second strategy for partial cleaning and a third strategy without data cleaning. In this way, by performing hierarchical cleaning on each logical block of the partition of the storage medium to be cleaned based on the first strategy, the second strategy and the third strategy, a targeted cleaning strategy is provided for different functional logical blocks of the storage medium to be cleaned, so that the data cleaning efficiency is improved while the data security is taken into account, and the user experience is improved.

The application belongs to the technical field of power monitoring, and discloses a power monitoring video fast splicing method based on key frame extraction, which comprises the following steps: video preprocessing, acquiring multiple power operation site monitoring videos, carrying out data cleaning and data enhancement, and laying a foundation for subsequent splicing; in the aspect of feature extraction network, the modeling capability of the algorithm on space and time sequence information is considered, a time shift module TSM is embedded into a two-dimensional convolution network ResNet101, a two-dimensional convolution network TSM-ResNet101 with space and time sequence information modeling capability is obtained, the network is used as the feature extraction network of the algorithm, non-overlapping areas are directly spliced, and panoramic key frames are generated; panoramic video generation: the panoramic key frames are reorganized according to the time sequence of the original video, a linear interpolation method is used to supplement the interframe transition effect, and a complete and smooth 360° panoramic monitoring video is generated. The application effectively resists electromagnetic interference in the power site, improves the quality of key frames, and provides protection for accurate registration.