12 results about "Background factors" patented technology

The invention discloses a geological disaster monitoring and early warning method based on multi-factor dynamic coupling, belongs to the technical field of geological disaster monitoring and early warning, and aims to solve the problems that a traditional method cannot fuse multi-source factors in real time, is low in early warning precision, lags in response and the like. A geological environment static background factor is combined to construct a susceptibility evaluation model, a dynamic weight is analyzed and calculated by adopting a time sequence, a dynamic Bayesian network is utilized to carry out coupling analysis, and a geological disaster risk probability value is output in real time, so that a corresponding early warning level and an emergency response are triggered. The method is mainly used for real-time monitoring, accurate risk assessment and timely early warning of geological disasters.

Provided is a technique for examining the intensity of difficulties related to the developmental characteristics of children and presenting the background factors and support directions of the difficulties and characteristics. 【Solution means】The server device includes a classification score calculation unit 11d-1 that obtains classification elements based on the answers to the corresponding question items for each sub-classification, calculates a classification score based on the classification elements, determines the intensity of difficulties for each sub-classification based on the classification score, and determines the intensity of difficulties for each major classification; a background calculation unit 11d-2 that calculates background elements for each background from the scores of the question items corresponding to the background, calculates a background score corresponding to the background elements, and performs a display setting of the background to be presented to the user based on the background score; and a support calculation unit 11d-3 that calculates the total score of the background scores of the backgrounds corresponding to the support as the support elements of each support, calculates a support score corresponding to the support elements, and performs a display setting of the support to be presented to the user.

The invention provides a small target detection method for a working image of a gantry crane, which belongs to the technical field of image processing, and comprises a crane working image data set construction method, aiming at the special working environment of a seaport gantry crane, rectangular anchor frames with different rotation angles are adopted to accurately mark each object, and the detection accuracy is improved. Therefore, the recognition accuracy is improved; according to the YOLOv5OBB optimization model based on the attention mechanism, a Center-BRA attention module applying global information and local information is designed, in order to make full use of space information and information between channels, a mixed attention branch is further added in the Backbone stage of the model, the lightweight Spatial Group-wise Enhance attention mechanism is introduced into a C3 module of a Neck part, important features are further enhanced, and the attention mechanism is optimized. Interference caused by noise or other background factors is ignored; based on a loss function Cscale Loss of target size information, the model pays attention to a detection frame which is smaller in size or is more difficult to correctly classify.

The invention relates to the technical field of natural language processing and medical artificial intelligence, and discloses a large model optimization method and system fusing a medical knowledge graph, and the method comprises the steps: constructing a time fuzziness-confidence joint modeling network model, time information corresponding to a current chief complaint symptom, context chief complaint content and patient individual information are input, a normal estimation time prediction result is output, joint calibration is conducted on the normal estimation time prediction result based on causal path information in the medical knowledge graph, and time prior distribution on the graph side is obtained; and constructing joint distribution based on the normal estimation time prediction result and the time prior distribution at the map side, taking the expectation of the joint distribution as a time point prediction result, inputting the time point prediction result into a large model, outputting a structured suggestion, and feeding back the structured suggestion to the medical knowledge map for updating. Therefore, the problem that inference is wrong due to the fact that semantic dependence cannot be recognized or patient background factors are ignored is solved.

To provide a technique for inspecting the intensity of troubles related to development characteristics of children, and presenting the troubles, background factors of the characteristics, and directionality of support.SOLUTION: A server device comprises: a classification score calculation section 11d-1 which obtains a classification raw score based on answers of question items corresponding to small classification, calculates a classification score based on the classification raw score, discriminates an intensity of troubles for each small classification using the classification score, and discriminates an intensity of troubles for each large classification; a background calculation section 11d-2 which calculates a background raw score for each background from scores of question items corresponding to the background, calculates a background score corresponding to the background raw score, and performs display setting of the background to be presented to a user based on the background score; and a support calculation section 11d-3 which calculates a total score of the background score of the background corresponding to support, as a support raw score of each support, and calculates a support score corresponding to the support raw score, and performs display setting of the support to be presented to the user.SELECTED DRAWING: Figure 2

The present invention discloses a small sample learning image classification method based on meta-features, and belongs to the field of small sample learning image classification. It is used in scenes with only a small number of labeled image samples, and can well complete classification tasks. Since there will be interference from background factors when measuring the similarity between image feature representations, the classification accuracy is poor. At the same time, in fine-grained tasks, it is not easy to distinguish between two samples of similar categories due to the small sample size. Therefore, based on this factor, the present invention proposes a small sample learning method based on meta-features, that is, when learning to obtain image feature representations, global feature representations and local feature representations are taken into account at the same time, and an adaptively adjusted network is established, so that the network model can adjust the attention to the global features and local features of the image according to the task in coarse and fine-grained tasks. Experiments have shown that the method discovered in this invention can achieve better results than traditional methods.

The invention discloses a water conservation and ecological restoration area identification method and device. The method comprises the following steps: firstly, obtaining the water conservation amount and water conservation influence factors of a target area; screening and detecting the water conservation amount and the water conservation influence factors to obtain an optimal water conservation value, a dominant factor and a background factor; clustering, classifying and superposing the background factors, and performing partition processing on the background factors and the dominant factors to obtain a plurality of effective ecological partition units; the optimal water source conservation value is combined to count and quantify the water source conservation amount of each effective ecological partition unit, and a water source conservation potential value and a potential surplus amount are obtained; and finally, according to the water conservation potential value and the potential surplus, dividing a restoration priority level, and identifying a restoration area. According to the method, the defects that a traditional method is low in recognition accuracy and weak in priority pertinence are overcome, the zoning and evaluation scientificity is improved, and the water conservation and ecological restoration efficiency is effectively improved.

The application discloses a water source conservation and ecological restoration area identification method and device. The method comprises the following steps: first, obtaining the water source conservation amount and water source conservation influencing factors of a target area; screening and detecting the water source conservation amount and the water source conservation influencing factors to obtain optimal water source conservation values, dominant factors and background factors; clustering and classifying the background factors and the dominant factors after superimposing the background factors to perform partition processing on the background factors and the dominant factors, thereby obtaining a plurality of effective ecological partition units; combining the optimal water source conservation values to quantify the water source conservation amount of each effective ecological partition unit, thereby obtaining water source conservation potential values and potential surplus amounts; and finally, dividing restoration priority levels according to the water source conservation potential values and the potential surplus amounts to identify restoration areas. The application makes up for the problems of low identification accuracy and weak priority targeting of traditional methods, improves the scientificity of partition and evaluation, and effectively improves the water source conservation and ecological restoration efficiency.

The present invention relates to the fields of natural language processing and medical artificial intelligence technology, and discloses a large-scale model optimization method and system that integrates a medical knowledge graph. The method includes: constructing a temporal fuzziness-confidence joint modeling network model, wherein the input includes the time information corresponding to the current chief complaint, the context of the chief complaint, and individual patient information; the output is a normal estimated time prediction result; the normal estimated time prediction result is jointly calibrated based on the causal path information in the medical knowledge graph to obtain the time prior distribution on the graph side; a joint distribution is constructed based on the normal estimated time prediction result and the graph-side time prior distribution, and the expectation of the joint distribution is used as the time point prediction result. The time point prediction result is input into the large-scale model to output structured suggestions, which are then fed back into the medical knowledge graph for updating. This solves the problem of reasoning errors caused by failing to identify semantic dependencies or ignoring patient background factors.

The application discloses a wheat ear detection method based on an improved YOLOv8s, which is suitable for improving wheat ear detection problems in a complex field environment, so as to realize rapid identification and effective counting of wheat ears. First, the application replaces a C2f module in a Backbone network by using a FasterBiNetBlock, which can reduce model parameters, reduce the number of calculation redundancies and memory accesses, and refine features of the model, so that more attention is paid to the wheat ears during detection, and interference of background factors such as wheat stems and wheat leaves is inhibited. Secondly, a new network structure EQFPN is proposed to strengthen a Neck network, so that the model has the characteristics of being friendly to quantization and hardware. Finally, a triplet attention module is introduced to capture cross-dimension dependency and wheat ear features, and meanwhile, problems such as false detection and missed detection during wheat ear detection are reduced.

The application discloses a kind of heterogeneous compact reservoir difference evolution model construction methods, belong to oil and gas geological exploration technical field, this method includes five core steps: first, core, experiment, well logging and seismic multi-source data are standardized and preprocessed;Then quantitative analysis reservoir porosity, permeability temporal and spatial variation and pore throat characteristics;Then dynamic source rock diagenesis process and quantize its intensity;Further, key factors such as sedimentary microfacies, diagenetic transformation intensity, tectonic activity frequency and burial depth are coupled, and the correlation function between them and reservoir densification degree is established;Finally, a visual model is constructed and optimized by validation data and regional geological background factors.The application solves the problems of multi-source data integration, evolution law quantization and poor model applicability, realizes the integrated dynamic characterization from micro-pore throat to macro-reservoir, significantly improves the prediction accuracy of high-quality reservoir, and provides reliable technical support for compact oil and gas reservoir exploration and development.

The invention discloses a circuit board welding fault identification method and system based on machine vision, and relates to the technical field of machine vision and image processing. Performing spatial position weighting on the high-resolution features by using background suppression guide information from a large receptive field branch, so that the model maintains small target texture and edge information and reduces the response ratio of a non-welding spot region, thereby improving the detectability of micro welding spots and fine-grained defects under a complex background; according to the method, a large receptive field feature extraction branch is set to provide a welding spot contour and context relationship, and welding spot enhancement guide information is generated by a high-resolution branch to participate in a feature fusion process of a second branch, so that the second branch adopts differential fusion weights for a welding spot candidate region and a background region in a fusion stage; and interference of background factors such as light reflection, silk screen and board texture on large target regression and classification is reduced.