Abstract: The present specification discloses video matching. In a computer-implemented method, a plurality of feature vectors of a target video is obtained. A candidate video similar to the target video is retrieved from a video database based on the plurality of feature vectors of the target video. A time domain similarity matrix feature map is constructed between the target video and the candidate video based on the target video and the candidate video. Using the time domain similarity matrix feature map as an input into a deep learning detection model, a video segment matching the target video in the candidate video and a corresponding similarity is output.

Abstract: The present specification discloses video matching. In a computer-implemented method, a plurality of feature vectors of a target video is obtained. A candidate video similar to the target video is retrieved from a video database based on the plurality of feature vectors of the target video. A time domain similarity matrix feature map is constructed between the target video and the candidate video based on the target video and the candidate video. Using the time domain similarity matrix feature map as an input into a deep learning detection model, a video segment matching the target video in the candidate video and a corresponding similarity is output.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for blockchain-based storage of patrol inspection proof are provided. One of the methods includes: receiving proof information of patrol inspection at a patrol inspection site; obtaining one or more operating environment parameters of the mobile terminal at a time of the patrol inspection, wherein the one or more operating environment parameters comprise at least location information and sensor data of the mobile terminal; performing credibility verification on the patrol inspection site based on the proof information, the location information, and the sensor data using an algorithm model based on labeled identification information, labeled operating environment parameters, and historical attendance information; and in response to the credibility verification being successful, uploading verification information associated with the proof information for storing in a blockchain.

Abstract: A computer-implemented method, non-transitory, computer-readable medium, and computer-implemented system are provided for detecting text similarity. A smart contract for detecting a similarity to a target original text is deployed in a blockchain network. The method is performed by a network node in the blockchain network. A first transaction, including first text, is received, where the first text is text whose similarity to the target original text is to be detected. The smart contract is invoked and text similarity detection logic stated in the smart contract is executed to obtain a detection result of similarity between the first text and the target original text.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for blockchain-based storage of patrol inspection proof are provided. One of the methods includes: receiving proof information of patrol inspection at a patrol inspection site; obtaining one or more operating environment parameters of the mobile terminal at a time of the patrol inspection, wherein the one or more operating environment parameters comprise at least location information and sensor data of the mobile terminal; performing credibility verification on the patrol inspection site based on the proof information, the location information, and the sensor data using an algorithm model based on labeled identification information, labeled operating environment parameters, and historical attendance information; and in response to the credibility verification being successful, uploading verification information associated with the proof information for storing in a blockchain.

Abstract: A computer-implemented method, non-transitory, computer-readable medium, and computer-implemented system are provided for detecting text similarity. A smart contract for detecting a similarity to a target original text is deployed in a blockchain network. The method is performed by a network node in the blockchain network. A first transaction, including first text, is received, where the first text is text whose similarity to the target original text is to be detected. The smart contract is invoked and text similarity detection logic stated in the smart contract is executed to obtain a detection result of similarity between the first text and the target original text.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for blockchain-based storage of patrol inspection proof are provided. One of the methods includes: receiving proof information of patrol inspection at a patrol inspection site; obtaining one or more operating environment parameters of the mobile terminal at a time of the patrol inspection, wherein the one or more operating environment parameters comprise at least location information and sensor data of the mobile terminal; performing credibility verification on the patrol inspection site based on the proof information, the location information, and the sensor data using an algorithm model based on labeled identification information, labeled operating environment parameters, and historical attendance information; and in response to the credibility verification being successful, uploading verification information associated with the proof information for storing in a blockchain.

Abstract: Implementations of the present specification provide an interaction behavior detection method, apparatus, system, and device. The method includes the following: obtaining a to-be-detected depth image photographed by a depth photographing device, extracting a foreground image used to represent a moving object from the to-be-detected depth image, obtaining spatial coordinate information of the moving object based on the foreground image, comparing the spatial coordinate information of the moving object with spatial coordinate information of a shelf in a rack, and determining an article touched by the moving object based on a comparison result and one or more articles on the shelf.

Abstract: A computer-implemented method includes: obtaining, by a detection device, first device information of a first set of devices detected in a target area using a first communication mode; determining, based on the first device information, a first quantity of devices in the target area; determining, based on second device information of a second set of devices, a verification coefficient, the second set of devices being detected in the target area using a second communication mode; and calculating, based on the first quantity of devices and a verification coefficient, a measure of real-time human traffic in the target area.

Abstract: A computer-implemented method, non-transitory, computer-readable medium, and computer-implemented system are provided for detecting text similarity. A smart contract for detecting a similarity to a target original text is deployed in a blockchain network. The method is performed by a network node in the blockchain network. A first transaction, including first text, is received, where the first text is text whose similarity to the target original text is to be detected. The smart contract is invoked and text similarity detection logic stated in the smart contract is executed to obtain a detection result of similarity between the first text and the target original text.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for blockchain-based storage of patrol inspection proof are provided. One of the methods includes: receiving proof information of patrol inspection at a patrol inspection site; obtaining one or more operating environment parameters of the mobile terminal at a time of the patrol inspection, wherein the one or more operating environment parameters comprise at least location information and sensor data of the mobile terminal; performing credibility verification on the patrol inspection site based on the proof information, the location information, and the sensor data using an algorithm model based on labeled identification information, labeled operating environment parameters, and historical attendance information; and in response to the credibility verification being successful, uploading verification information associated with the proof information for storing in a blockchain.

Abstract: Disclosed are methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing interaction behavior detection. One method includes obtaining a depth image from a depth camera disposed above a rack, and a filming range of the depth camera covers an outer edge of each shelf in the rack and an aisle. A foreground images associated with a moving object from the depth image is extracted. Spatial coordinates of the moving object are obtained based on the foreground image. An object on a shelf in the rack that is physically contacted by the moving object is determined based on comparing the spatial coordinates of the moving object with spatial coordinates of the shelf, where the spatial coordinates of the shelf is determined based on a background depth image captured by the depth camera at a predetermined angle unobstructed by the moving object.

Abstract: A computer-implemented method, non-transitory, computer-readable medium, and computer-implemented system are provided for detecting text similarity. A smart contract for detecting a similarity to a target original text is deployed in a blockchain network. The method is performed by a network node in the blockchain network. A first transaction, including first text, is received, where the first text is text whose similarity to the target original text is to be detected. The smart contract is invoked and text similarity detection logic stated in the smart contract is executed to obtain a detection result of similarity between the first text and the target original text.

Abstract: The present invention provides a method and apparatus for scheduling block device input/output requests, which relates to the field of computer memories. The method comprises: generating a block device input/output request firstly, wherein the block device input/output request comprises a write operation request; then judging whether the generated block device input/output request can be combined with block device input/output requests in a request queue; and if the generated block device input/output request cannot be combined with block device input/output requests in a request queue, sending the generated block device input/output request to a foremost position of the request queue when a destination device of the generated block device input/output request is a solid state disk or an SD card. By means of the method and apparatus for scheduling the block device input/output requests, the efficiency of write operation is improved.

Abstract: The present invention discloses a method and a device for generating a CNM, so as to improve a processing rate of a congestion problem. The method includes: monitoring a data volume of an output queue at a congestion point CP; monitoring a rate of a data stream corresponding to at least one congestion reaction point RP connected to the CP, and when the data volume of the output queue at the CP reaches a congestion threshold, obtaining identification information of a data stream that currently has a highest rate; generating a CNM for the data stream having the identification information, and sending the CNM to an RP corresponding to the identification information, so that the RP reduces, according to the CNM, the rate of the data stream corresponding to the RP.

Abstract: The present invention provides a method and apparatus for scheduling block device input/output requests, which relates to the field of computer memories. The method comprises: generating a block device input/output request firstly, wherein the block device input/output request comprises a write operation request; then judging whether the generated block device input/output request can be combined with block device input/output requests in a request queue; and if the generated block device input/output request cannot be combined with block device input/output requests in a request queue, sending the generated block device input/output request to a foremost position of the request queue when a destination device of the generated block device input/output request is a solid state disk or an SD card. By means of the method and apparatus for scheduling the block device input/output requests, the efficiency of write operation is improved.

Abstract: The present invention provides a method and device for organizing and restoring file indexes, which belongs to a technical field of data storing methods. In the present invention, a circular bidirectional linked list is constituted by all files in a storage medium firstly, wherein the bidirectional linked list comprises a forward linked list and a backward linked list; the indexes of all files are stored in an index information set, and each file is provided therein with a backup of its corresponding index; afterwards, in the process of accessing of a file, if the index of the file in the index information set doesn't match with the backup of the index of the file, the bidirectional linked list is restored; finally, the indexes of all files are re-established according to the restored bidirectional linked list. By using the method and device for organizing and restoring file index of the present invention, an efficiency of file index reestablishment can be improved.

Abstract: The present invention provides a method and device for restoring system file indexes, which belongs to the technical field of data store methods. In the present invention, when a storage medium is loaded and an index information set is not read, or when the storage medium is loaded and the indexes in the index information set are judged to be incorrect, or when a read operation is executed for a file in the storage medium and an index of the file in the index information set doesn't correspond to a backup index in the file, or when a writing operation is executed for a file in the storage medium and an index of the file in the index information set doesn't correspond to a backup index in the file, the backup indexes of all files in the storage medium are updated, and the index information set is restored or re-established. By using the method and device for restoring system file indexes of the present invention, a deficiency of repeated restarts without recording is overcome.

Abstract: The present invention discloses a method and a device for generating a CNM, so as to improve a processing rate of a congestion problem. The method includes: monitoring a data volume of an output queue at a congestion point CP; monitoring a rate of a data stream corresponding to at least one congestion reaction point RP connected to the CP, and when the data volume of the output queue at the CP reaches a congestion threshold, obtaining identification information of a data stream that currently has a highest rate; generating a CNM for the data stream having the identification information, and sending the CNM to an RP corresponding to the identification information, so that the RP reduces, according to the CNM, the rate of the data stream corresponding to the RP.

Abstract: Candidate promotion keywords are selected. Features of the candidate promotion keywords are extracted. The features include at least one of a search engine feature, an effect feature of non-directed traffic, and a text feature. The features of the candidate promotion keywords are used as input data of a pre-established keyword screening model, and superior promotion keywords are obtained according to a prediction result of the keyword screening model.