12results about How to "Efficient implementation" patented technology

This invention discloses a method and system for constructing a vectorized feature set of an indoor scene point cloud model. The method includes acquiring point cloud data, visible light images from different directions, and depth maps for the indoor scene; performing target object segmentation on the point cloud data based on semantics, and image segmentation on the visible light images and depth maps based on edges; extracting 3D features from enclosed regions of the point cloud, visible light images, and depth maps based on the segmentation results; and then fusing them to generate a vectorized representation of the spatial 3D visual data of the indoor scene. This invention is simple to implement and can achieve a complete expression of multi-dimensional scene data based on point cloud, visible light, and depth map data. Due to the use of a graph structure for data vectorization and fusion, the computational load can be effectively reduced, enabling efficient and low-cost construction of a vectorized feature set for an indoor scene point cloud model.

This invention belongs to the field of drilling stability analysis technology, specifically a wellbore stability evaluation method, electronic equipment, and media. This method establishes a wellbore stability index for the first time, using the average of the wellbore stability indices from multiple wellbore stability sample wells as a threshold. The closer the target well's wellbore stability index is to the threshold, the better the wellbore stability; conversely, the further away, the worse the wellbore stability. This invention's method can evaluate wellbore stability while drilling with high accuracy. The electronic equipment and media based on this invention broaden its application scenarios, making it suitable for evaluating the wellbore stability of horizontal wells in sandstone and mudstone formations.

The application relates to a target detection simulation method based on a satellite-borne load, which comprises the following steps: satellite orbit simulation, star background simulation, visible target judgment and detection image simulation; image processing is performed on the ideal detection image so that the detection image simulation result approaches the detection image of the actual satellite-borne load. The method is realized by using C++ programming to improve the algorithm safety and flexibility.

The application discloses application of Neuritin in preparation of a medicine for promoting healing of a diabetic wound. The application proves for the first time through experiments that Neuritin expression in wound edge tissue and surrounding sensory neurons is significantly up-regulated after skin damage, and Neuritin expression is significantly down-regulated under a diabetic condition. By constructing a nervous system conditional knock-out and a sensory neuron specific knock-out Neuritin mouse, it is proved that wound healing is significantly delayed after Neuritin is knocked out. By means of spinal cord site injection of an AAV virus to overexpress Neuritin in a DRG and skin electroporation technology to locally overexpress Neuritin in the skin, it is proved that overexpression of Neuritin can significantly promote wound healing of type 1 and type 2 diabetic mice. The application provides a brand-new target and strategy for treatment of a diabetic refractory wound, and has a good clinical transformation prospect.

This invention belongs to the field of industrial internet security technology, specifically relating to an industrial internet vulnerability verification method that addresses the problems of low efficiency, insufficient accuracy, poor engineering feasibility, and inconvenient remediation in existing vulnerability verification technologies. This invention first constructs a protocol template library and parses the original network packets to obtain field sequences; it calculates the distribution entropy, historical mutation response rate, and type identifier as scenario feature vectors; it retrieves similar historical scenarios and dynamically selects strategies based on mutation strategy statistics and experience scores; it generates test cases from mutation seed packets, sends them to the target device for verification, and obtains responses; it updates strategy statistics and the experience replay pool; in case of anomalies, it actively intervenes in experiments to determine causal effects, generating verification records, reproduction packages, and guidelines for manual confirmation. This invention achieves accurate and efficient vulnerability verification, improves the interpretability and reproducibility of results, reduces engineering deployment costs, and facilitates vulnerability analysis and remediation.

The application discloses a kind of higher education credit point automatic accounting and early warning system and method, including educational administration data acquisition and standardization module, training scheme and point rule analysis module, credit point automatic accounting module, academic risk assessment and early warning module, early warning information push and feedback module, data archiving and dynamic updating module;The application is replaced artificial accounting process by multi-source data standardization and automatic accounting model, eliminates human error, realizes the full-process automation of point accounting, and the accounting caliber is unified, the result is accurate, adapts the differentiated conversion needs of difference of conventional course and special course.Cultivation scheme and accounting rule analysis storage mechanism are used, the decoupling of rule and business logic is realized, and the differentiated requirements of different majors, different grades and different education levels can be quickly adapted, and the rule is updated conveniently, and the system has strong expansibility.

The utility model relates to the technical field of industrial operation and maintenance, aims at solving the problem that the hanging rail type inspection robot cannot return to the charging pile autonomously after being fed with power at non-charging points, and manual traction is needed for the maintenance personnel to work at high places, and there are safety risks, interruption of inspection, low operation and maintenance efficiency and increased cost, provides an emergency charging interface suitable for track robots, including robot, the robot has the temporary emergency charging mouth, and the temporary emergency charging mouth and the battery inside the robot are mutually parallelly connected, and the temporary emergency charging mouth is detachably connected with the connecting cable, and the connecting cable is provided with the power adapter, the utility model has the beneficial effect that the safety risk of working at high places is completely eliminated, the operation and maintenance efficiency and usability are improved, the system reliability is enhanced, the minimum modification and low -cost deployment are realized, the human resource allocation is optimized, the equipment service life is prolonged and the operation is simple and easy to implement.

This invention relates to a method and system for defending against backdoor attacks in federated learning, belonging to the field of information security. The method first performs data preprocessing operations on the terminal before executing local training, converting the data from the time domain to the frequency domain and designing a low-pass filter to filter high-frequency signals introduced by backdoor samples. Second, malicious uploads and updates are detected on the server side, the cosine similarity between the pre-global model and the local model is calculated, and an appropriate threshold is set to filter maliciously uploaded local model updates. Finally, a global aggregation operation is performed on the remaining non-malicious uploads and updates to obtain an updated global model, and differential privacy technology is used to perturb the obtained global model. This invention fully considers the scenario where backdoor attacks occur in different locations, mitigating backdoor attacks in federated learning without affecting the model's performance on benign data, thus solving the problem of backdoor attack defense in federated learning.

The application discloses a system for purifying synthetic gas and regenerating rapeseed oil methyl ester, which regenerates the rapeseed oil methyl ester by removing impurities and tar by means of the steam method of a stripping unit, and adds the regenerated rapeseed oil methyl ester into a washing unit, so that the rapeseed oil methyl ester in the washing unit can better absorb the tar under the condition that a small amount of fresh rapeseed oil methyl ester is added into the washing unit, and the phenomenon of supersaturation is reduced as much as possible; the application also uses the regenerated rapeseed oil methyl ester and the mixed emulsion after cooling to make the moisture in the synthetic gas into condensed liquid through condensation, and capture the water-soluble pollutants of the synthetic gas through the condensed liquid; in conclusion, the application can remove the tar and the water-soluble pollutants at the same time under mild conditions, and can efficiently utilize the rapeseed oil methyl ester.

This application discloses a method, apparatus, device, medium, and product for cognitive function assessment, relating to the fields of computer and speech processing technology. The method includes: acquiring target speech data; extracting features from the target speech data according to feature categories within a preferred subset of speech features corresponding to the target cognitive test task, obtaining a target speech feature set; inputting the target speech feature set into a cognitive function assessment model corresponding to the target cognitive test task, obtaining the cognitive function assessment result of the target test subject; wherein, by constructing a multimodal cognitive database, extracting wide-area speech features from each form of speech data in the database, and based on the wide-area speech features and corresponding cognitive function assessment results, selecting a preferred subset of speech features based on the correlation strength with cognitive function, and determining the cognitive function assessment model, machine learning and statistical methods are used. This application achieves accurate and efficient cognitive function assessment.

This application provides a CSI-RS transmission method, apparatus, and device. The method includes: receiving resource configuration information corresponding to a CSI-RS resource set sent by a base station; wherein the resource configuration information is used to indicate the CSI-RS resource set, and the CSI-RS resource set includes CSI-RS resources located in an SBFD symbol; wherein the CSI-RS resources exceed the downlink subband range of the SBFD symbol, or the CSI-RS resources are located in the downlink subband of the SBFD symbol; receiving CSI-RS sent by the base station based on the CSI-RS resources in the CSI-RS resource set, measuring the received CSI-RS, and reporting the measurement results to the base station. This solution can improve resource utilization.