10results about How to "Creative" patented technology

The present application belongs to the technical field of ground laser radar forestry data processing, and discloses a broad-leaved forest single tree segmentation method and system based on branch information guidance, taking ground-based laser radar broad-leaved forest point cloud as a processing object, using RANSAC cylindrical fitting to combine the growth characteristics of the tree trunk to detect the tree trunk; starting from the top of the tree trunk in the low vegetation area, the tree branches are extracted by segmenting and growing the branches and combining the thickness changes of the branch segments; starting from the end of the branch, the tree crown leaf point cloud is segmented by layer-by-layer growth. The present application has stronger trunk detection capability under the conditions of lush low vegetation and complex terrain; on the other hand, the present application can accurately segment the tree crown when the large and small crowns are intertwined and the multiple crowns are closely surrounded. The algorithm is efficient, simple and easy to use, and has great practical significance for improving the semantic understanding ability of forest scenes, assisting forest resource investigation, vegetation ecological research, and satellite remote sensing product calibration and verification.

This invention belongs to the field of time measurement technology and discloses a resource-saving short-chain time-to-digital converter and its conversion method. It utilizes a delay unit and conversion circuit to build a local oscillator, which flips the output state at fixed time intervals during a continuous high-level input signal. The time at which the local oscillator causes the level to flip is known. The time of the last flip is obtained using TDL and TDC, and the start and last flip times are obtained by recording the number of flips of the local oscillator. The resource-saving short-chain time-to-digital converter provided by this invention has good linearity, completely avoiding nonlinearity problems caused by inconsistent line lengths introduced across multiple resource blocks. The resource-saving short-chain time-to-digital converter has a simple and flexible structure, low resource consumption, and high utilization rate. It has good robustness and can be directly ported between different channels or even devices, and can correct temperature-induced drift online.

The application belongs to the technical field of signal transmission, and discloses a method for eliminating analog signal transmission error through digital compensation, which specifically comprises the following steps: in addition to the original signal transmitting circuit part, a signal detecting part composed of MCU, ADC and a communication module is arranged at the signal transmitting end; a signal detecting part composed of MCU and ADC communication module and a control part capable of controlling the signal at the output end to be offset and gain varied are arranged at the signal receiving end; the signal at the receiving end is collected by the MCU to compare with the data of the transmitting end communication module, and the difference between the two data is calculated to obtain the correction value of the amplitude and offset; the gain and offset of the signal are changed by controlling the offset and gain circuit part at the receiving end, so that the transmission error is eliminated, and the gain and offset control method is realized by using DAC, PGA and other components; the deviation of the frequency response can also be calculated according to the requirement, and is corrected by the gain and offset control circuit part.

The application belongs to the technical field of trusted authentication, and discloses a remote user trusted authentication system, method and medium in an Internet of Things environment. The system comprises a registration center (RC), which is a trusted server and is used for safely registering or initializing other entities; an intelligent device, which is arranged in an environment and is used for collecting and transmitting data; a gateway, which supports interconnection between different protocol networks and has computing and storage capabilities; a user, who accesses remote intelligent devices through a login device; and the login device, which is used for user trustworthiness verification. The login device includes but is not limited to a mobile device, an intelligent card and a PC. Through the system and method, the trusted authentication of remote users can be realized, and safe communication between the user and the intelligent device is established. Meanwhile, the application can also solve the problems of public key cryptography complexity, user authentication vulnerability and symmetric cryptography limitations.

This invention provides a method and apparatus for deep inference of an archipelago through computing power on an intelligent computing center cloud platform. It relates to the fields of intelligent computing centers, smart computing centers, computing power infrastructure, and smart computing cloud technologies. The method includes: Step S1, generating N initial response data corresponding to an initial problem, where N is a positive integer greater than 1; Step S2, dividing the N initial response data into M islands, each island including K initial response data, where M and K are positive integers greater than 1; Step S3, evolving the K initial response data included in each island to obtain K evolved response data included in each island; Step S4, generating first inference response data based on the K evolved response data of each of the M islands; Step S5, outputting the first inference response data. This invention can significantly improve the accuracy of the response.

This invention belongs to the field of artificial intelligence algorithm technology and discloses an intelligent transfer adversarial method and system for electromagnetic signal recognition under incomplete information. First, multiple models of alternative target networks with incomplete information are constructed in parallel, and these alternative models are trained using an evolutionary strategy based on a prior knowledge matrix. Then, an ensemble network model is generated based on the trained alternative models, and a fully connected layer and a layer with feature-level weighted fusion are designed. Finally, an optimized adversarial method for electromagnetic spatial signal recognition tasks is designed, generating adversarial samples under the ensemble model, and transferring these adversarial samples to the target network to achieve adversarial capabilities against the target network. This invention can effectively achieve adversarial capabilities against the target network when the target network information is unknown, and it still exhibits good transfer adversarial performance under the same signal recognition task and different unknown target network conditions.

The application belongs to the technical field of transformers, and discloses a linear phase-shifting transformer core loss calculation method, system, device and terminal, which is based on a classical trinomial loss separation model, three loss coefficients are calculated by using a least square method; the influence of local hysteresis loop, magnetic field non-sine, tooth slot effect and skin effect on the loss model and the loss coefficient is analyzed, the static magnetic density characteristics of the linear phase-shifting transformer are obtained based on finite element simulation calculation; the tangential magnetic density and the radial magnetic density of each characteristic point are separated, and Fourier analysis is carried out, the fundamental wave magnetic density and each harmonic magnetic density are substituted into the improved loss separation model. The core loss of the linear phase-shifting transformer in the stable operation state can be accurately calculated, the calculation precision is greatly improved, which can provide a reference for the efficient operation of the transformer, and has important significance for the optimization design of the linear phase-shifting transformer body and the realization of the minimum loss control of the multiple superposition inverter system of the linear phase-shifting transformer.

This utility model provides a foldable traverse module assembly, comprising: a foldable truss module, a foldable stabilizing module, and a lighting module. The foldable truss module includes a foldable left cantilever, a foldable right cantilever, and a load-bearing truss. The foldable left and right cantilever are connected to the load-bearing truss via stepped shaft bolts and are locked and unlocked using U-shaped locking pins. This foldable traverse module assembly has the advantages of being easy to fold, easy to operate, portable, space-saving, and saving physical and manpower. It can promote the construction of outdoor sports, adventure, and rope traverse rescue equipment systems and improve water rescue capabilities. The industry market demand potential is huge, and it can be equipped by various outdoor sports and adventure teams, fire rescue teams, and professional water rescue teams. It has broad development prospects and is of great significance to public fire safety.

The application belongs to the field of computer vision, and discloses a weakly supervised semantic segmentation method, system, device and medium based on random combination, which comprises the following steps: training classification networks N1, N2 and N3 respectively by using a training data set, a slice training data set and a slice training data set randomly combined, so that each network can extract different active regions in the picture, and the learning results of the other two networks are learned by using the mutual supervision training mode; finally, the prediction results of the three networks are combined to obtain the final semantic segmentation result, which is used as a semantic segmentation training data set to train a semantic segmentation model to predict the final semantic segmentation result. The application effectively utilizes the different perception areas of the network for the randomly combined slice pictures, and utilizes the different classification networks to perceive the categories of the same picture, thereby improving the semantic segmentation ability and prediction accuracy of the semantic segmentation model through the semantic segmentation data set obtained by combining the results of the three classification networks.

The application belongs to the technical field of natural language processing, and discloses a federated learning optimization method and system for entity extraction, a device and a terminal. In the shared data meta information transmission stage, the server sends a shared data request to each selected client participating in federated learning, each client calculates shared data set meta information, and uploads the meta information to the server. In the approximate IID entity annotation data construction stage, the server constructs an approximate IID entity annotation data index set according to the meta information uploaded by the client, and requests the corresponding client for data corresponding to the index set. In the model training stage, the server is regarded as a client participating in federated learning training, and the approximate IID entity annotation data set is used for model training to obtain an approximate centralized model W IID The application reduces the difference between the model parameter spaces of the clients, improves the accuracy and convergence speed of the global model, and reduces the training communication cost.