8results about How to "Avoid repeated execution" patented technology

The invention discloses a voice text bidirectional conversion system and method based on asynchronous processing and a message queue. The system comprises a server, an OBS object storage service, a business system, a conversion gateway and a plurality of universal gateways. The conversion gateway is connected with the server, a voice-to-text module and a text-to-voice module are configured in the conversion gateway, and the voice-to-text module is constructed based on a FunASR streaming recognition framework, performs communication through a WebSocket protocol, and performs a voice recognition process in combination with an asynchronous task processing mechanism; the text-to-voice module is constructed based on ChatTTS service, and realizes an asynchronous task distribution and result return mechanism in combination with a RabbitMQ message queue. According to the scheme, efficient conversion between the voice and the text is realized through unified interface design and modular processing flow, and the response speed, the accuracy and the expandability of the voice processing system are improved.

The application discloses a query condition conversion method and device for adaptive AI output correction, and relates to the technical field of artificial intelligence and natural language processing. The method comprises the following steps: receiving a user natural language input, dynamically constructing a system prompt word based on target table field metadata, sending the system prompt word to a large language model together with the user natural language input, obtaining a JSON format query condition array and transmitting and storing the JSON format query condition array. JSON strings in the array are extracted and converted into a query condition object array, logical operator position correction processing is performed on the query condition object array, the first query condition logical relationship is set to be empty, and the logical operator value of the previous query condition is read as the logical relationship of the previous query condition. The corrected array is converted into a system internal data structure, a database query is performed by using the system internal data structure, and data records meeting the conditions are obtained and returned. The problem that a large language model sets a logical operator on a previous expression, resulting in incorrect query results, is solved.

The invention relates to the technical field of digital images, in particular to an image encryption and decryption method and device based on RISC-V, equipment, a storage medium and a computer program product. According to the method, an initial key is written into a coprocessor based on a preset extension instruction, and the internal state of the chaotic system is configured based on the initial key; based on the internal state of the chaotic system, sponge iteration is carried out, a random sequence is generated, and an image abstract is output; writing the storage position information of the plaintext image into a coprocessor, and enabling the coprocessor to read the plaintext image pixel by pixel in a direct storage access mode to obtain a pixel stream; in the coprocessor, performing reversible confusion operation and reversible diffusion operation on the sequence of the pixel flow to obtain a processed pixel flow; and in a decryption stage, an internal state which is the same as that of the chaotic system is constructed based on the image abstract, and the processed pixel stream is subjected to reverse operation according to the same hardware logic which is multiplexed in a reverse order, so that the original pixel stream is obtained, and the efficiency of image encryption and decryption is improved.

This invention provides a file hierarchical ciphertext policy attribute-based encryption method, system, and electronic device, relating to the field of cryptography. The method includes: a trusted center generating a master public key and a master private key; a user generating an identity private key pair, and the trusted center generating an outsourced decryption key based on the identity public key and sending it to a cloud server; the data owner constructing a hierarchical access tree, associating symmetric keys with level nodes, and employing a dual polynomial mechanism to allocate secret and noise components to each node, thereby generating a complete ciphertext containing hierarchical ciphertext, attribute ciphertext, and transmission ciphertext, which is then uploaded to the cloud server; after verifying the user's request, the cloud server preprocesses the complete ciphertext using the user's outsourced decryption key, obtaining an outsourced decrypted ciphertext returned to the user; finally, the user uses their identity private key to decrypt the outsourced decrypted ciphertext, recovering the symmetric key and obtaining the target plaintext. This invention maintains the computational efficiency of file hierarchical encryption while effectively resisting multi-user collusion attacks.

This invention discloses a method and system for medical image segmentation and 3D reconstruction based on region growing. The method includes the following steps: S1, acquiring the original medical image data of the target object and preprocessing it to generate a preprocessed 2D image sequence; S2, selecting seed points on the preprocessed 2D image sequence; S3, segmenting based on the seed points to generate segmented regions; S4, optimizing the boundaries of the segmented regions to generate an optimized binary mask; S5, generating the final binary mask; S6, completing the 3D model reconstruction; S7, spatially registering and fusing the reconstructed 3D model. The semi-automated segmentation process of this invention reduces the subjective differences of manual delineation, achieving over 90% consistency in reconstruction results for the same image across different operators, providing a reproducible technical basis for multi-center studies and efficacy evaluation.

The invention relates to a video coding method and device, electronic equipment, a storage medium and a program product. The video coding method comprises the following steps: sampling continuous frames in a video stream to be processed to obtain sampled frames and non-sampled frames; performing ROI (Region of Interest) detection on the sampling frame based on touch driving and / or content awareness, and identifying an ROI in the sampling frame; carrying out multiplexing or fine tuning on the ROI region through lightweight inter-frame difference analysis and ROI state judgment in a non-sampling frame; when video encoding processing is performed, the ROI region and the non-ROI region in the sampling frame and the non-sampling frame are encoded in a distinguished manner. According to the invention, the video quality of the key area can be guaranteed, and the overall calculation and coding overhead can be obviously reduced.

Embodiments of the present application provide a task processing method and device, a storage medium and a program product, and relate to the field of computers. The method comprises: in the case of adding a first task to a first queue, sending the first task to a first node matched with the first queue, storing a first key-value resource matched with the first task to a first storage space by the first node, and processing the first task based on the first key-value resource; in the case of the execution duration of the first task reaching a first predetermined duration, adding a second task matched with the first task to the first queue; in the case of determining that the execution duration of the first task reaches a second predetermined duration, sending the second task to the first node, storing a second key-value resource matched with the second task to the first storage space by the first node, and processing the second task based on a target resource in the second key-value resource and the first key-value resource. The above scheme solves the technical problem of wasting computing resources.

The application discloses a multi-tenant mode cross-platform data synchronization method and device, equipment and medium, and is used for solving the technical problem that the synchronization data task between multiple tenants is influenced mutually when being executed in series, resulting in poor synchronization effect. The application comprises the following steps: when a SaaS cloud platform generates an operation behavior, generating a data operation log table; determining a target tenant and an operation type corresponding to the operation behavior; generating a key-value pair of each target tenant, and generating a tenant latest key-value pair data format cache by using the key-value pair; creating an operation thread of each target tenant, and querying a target data operation log record from the data operation log table according to the key-value pair and the operation type of each target tenant; querying whether there is a data operation log lock mark of the corresponding target tenant; if not, creating a data operation log lock mark of the target tenant, and synchronizing operation data to a third-party SaaS cloud platform according to the target data operation log record.