8results about How to "Understand comprehensively" patented technology

The application discloses a brain glioma segmentation and three-dimensional visualization model training method and system based on multi-modal fusion, and comprises the following steps: inputting multi-modal medical image data into a Laplacian pyramid multi-modal fusion learning model to generate a fusion graph, then inputting the fusion graph into a U-net segmentation model comprising an encoder and a decoder to obtain a segmentation mask under the multi-modal fusion graph; and finally, based on the multi-modal fusion graph and the segmentation mask, adopting a Marching cubes (MC) algorithm to perform three-dimensional reconstruction visualization of brain regions and brain tumor region labeling. The brain glioma segmentation and three-dimensional visualization model training method and system based on multi-modal fusion provided by the application optimizes the disadvantages of insufficient information and incomplete modeling under single modal, fully utilizes the advantages of the completeness of multi-modal data, performs multi-modal fusion, segmentation and three-dimensional visualization, can more comprehensively assist surgeons in treatment, and improves the accuracy in the operation process.

The invention discloses a higher vocational nursing skill competition scoring system based on an attribute hierarchy model, and particularly relates to the field of education evaluation, and the system comprises a nursing skill index collection module, a nursing skill scoring hierarchy construction module, a nursing skill score collection module and a man-machine interaction module. According to the method, the nursing skills of the participating students are scored by establishing the nursing skill scoring hierarchical structure, the relative importance among the nursing skill scoring indexes is compared, the judgment matrix is established based on the nursing skill scoring hierarchical structure, and the index weight of the judgment matrix is calculated; a nursing skill comprehensive scoring model is established according to the index weight to obtain a nursing skill comprehensive score, and a scoring result can be fed back to participants through an attribute hierarchy model to help the participants to understand the performance in each attribute aspect, so that targeted improvement is carried out, and the experience of the participants is improved. Meanwhile, schools and teachers can comprehensively and objectively understand the performance of the competitors, and scoring deviation caused by subjectivity is avoided.

This invention relates to an address standardization processing system and method: It acquires all original address data related to a certain region; aggregates address data from the original address data that satisfy corresponding address rules; matches a standard address to a specific address data contained in the address rules; if a match is successful and the matched standard address is unique, it is manually reviewed and approved, and the address data is mapped to the standard address; if multiple standard addresses are matched, it is manually reviewed and a single standard address is finally determined and mapped; if the manual review fails, manual location is performed to verify the standard address mapped to the address data; if matching fails, manual location is notified; the standard address mapped to the address data is sent back to each original address data under that address data; the address data is extracted from each original address data and followed by detailed address data to supplement the corresponding standard address to obtain the detailed standard address corresponding to each original address data.

The embodiment of the invention relates to the technical field of plant management, and discloses an unmanned aerial vehicle-based vegetation AI identification management method, which comprises the following steps of: identifying wide-area remote sensing data to obtain all abnormal sub-areas and window detection areas in a target detection area; executing a second acquisition task on all abnormal sub-regions and window detection regions to obtain corresponding fine remote sensing data; determining vegetation state information of the abnormal sub-region according to the fine remote sensing data, determining a canopy shielding relationship in the window detection region according to the fine remote sensing data, and identifying and determining a potential under-forest region under canopy coverage; probabilistic inference is carried out on the vegetation state of the potential under-forest region through fusion of multi-source evidence, and an under-forest vegetation probability distribution diagram containing inferred vegetation categories and corresponding confidence coefficients is generated; and generating a comprehensive vegetation state report of the target detection area. And the overall identification management efficiency is improved through a layered acquisition mode.

The present disclosure relates to processing cell images in a convolutional neural network with contextual data. The present disclosure relates to image processing in machine learning convolutional neural networks. The technology can improve the accuracy and reliability of applications using machine learning convolutional neural networks, particularly in the field of cell microscopy. Contextual data is provided which indicates cell features determined globally for a scene, and this data is then taken into account in the inference of the machine learning convolutional neural network.

This application provides a risk warning method, apparatus, electronic device, and storage medium, relating to the field of risk management technology, for more flexible and effective risk warning and risk handling of risky business operations. The method includes: determining a business model associated with the target business when an operation request for a target business exists and the operation request meets preset risk conditions; wherein the business model indicates applicable businesses and their business logic information; the applicable businesses include the target business; determining a business dataset corresponding to the target business based on the business model and a preset knowledge base; wherein the preset knowledge base includes risk logs corresponding to various businesses; the risk logs include risky operations; and providing risk warnings for the operation request based on the business dataset.

The utility model discloses a navigation mark ship with swimmer warning function relates to navigation mark ship technical field, include: navigation mark ship body, the periphery of navigation mark ship body evenly distributed with a plurality of inductor group, the top of cylinder type support is installed with navigation mark light, the top of cylinder type support is connected with the rotating sleeve through the bearing sleeve, install camera on the rotating sleeve, the upper portion lateral wall of cylinder type support is installed with loudspeaker horn at equal intervals, drive assembly, loudspeaker horn and camera are in low -power consumption standby mode when not working, inductor group low frequency periodicity detects. The utility model discloses through evenly distributed with a plurality of inductor group at the periphery of navigation mark ship body, can detect swimmer close -in in time, through loudspeaker horn sends warning sound, reminds swimmer to keep away, effectively avoids swimmer to climb or close -in navigation mark ship, reduces the risk of accident occurrence, guarantees the safety of swimmer and navigation mark ship.

The terrain fusion method based on a digital elevation model and radar active detection of the application comprises the following steps: obtaining terrain point information detected by radar antenna rows, compensating terrain point information detected by multiple antenna rows to the same coordinate system, and saving the terrain point information in the same coordinate system according to a preset interval of azimuth angle; fitting the terrain point information; eliminating height deviation after fitting; terrain fusion; and generating a terrain elevation map in front of a current carrier to improve the use performance of the radar.