9results about How to "Good distinction" patented technology

The application discloses a method, system, chip, device and computer readable storage medium for rapidly identifying the producing area of Amomum villosum and evaluating the quality of Amomum villosum based on an electronic nose and an electronic tongue, and belongs to the technical field of traditional Chinese medicine identification. The technical problems to be solved are low detection efficiency and high cost of the existing Amomum villosum identification method. The technical solution points are as follows: a method for rapidly identifying the producing area of Amomum villosum and evaluating the quality of Amomum villosum based on an electronic nose and an electronic tongue, which comprises the following steps: electronic nose test sample preparation and electronic tongue test sample preparation; an electronic nose system adopts a direct headspace air suction method to collect signals of the electronic nose test sample; an electronic tongue system collects signals of the electronic tongue test sample; average values of response values are extracted to form an odor characteristic vector; data analysis and producing area identification / quality evaluation are performed according to the fusion characteristic vector, and finally, model application is performed.

The invention belongs to the field of cassava molecular markers, and particularly relates to a cassava KASP marker and application of the cassava KASP marker in cassava variety distinguishing. The 27 KASP markers provided by the invention have good distinguishing ability in cassava genetic analysis, and can effectively reveal genetic differences among different materials. The DNA fingerprint spectrum of 184 parts of cassava materials is constructed based on 27 core KASP markers, and the purpose of rapid variety identification can be achieved. The KASP marker disclosed by the invention can be used for effectively identifying false selfing offspring. In conclusion, the KASP marker provided by the invention provides a new method and thought for genetic improvement and variety identification of cassava, provides a powerful tool for subsequent cassava breeding and genetic improvement research, and is beneficial to breeding of new varieties with excellent characters.

This invention discloses an image-aware hashing method and system based on residual networks, belonging to the field of information security; it includes S1, using a residual network to perform a first-stage training on an image group to obtain a hash sequence; S2, reading the weights after the completion of the first-stage training, using these weights to perform a second-stage training, and applying a loss function L... f2 By adjusting the network to increase the distance between the original image and the approximate image, the network is ultimately able to distinguish between the original image and the approximate image, thus overcoming the shortcomings of existing technologies.

This invention relates to the field of biomedical technology and proposes RBP4 and Ferritin as biomarkers for evaluating the efficacy of hormone therapy for allergic rhinitis. The biomarkers include one or a combination of RBP4 and Ferritin. Studies have confirmed that both RBP4 and Ferritin are capable of distinguishing between hormone-sensitive and hormone-insensitive allergic rhinitis. RBP4 is expressed at higher levels in the hormone-sensitive allergic rhinitis group, while Ferritin is expressed at higher levels in the hormone-insensitive allergic rhinitis group. Compared to single biomarker detection, the combined use of RBP4 and Ferritin significantly improves the differential diagnostic efficacy, providing better differentiation between the two patient groups.

The invention provides a brocade pattern generation method and system based on a diffusion model, and relates to the technical field of pattern generation, and the method comprises the steps: collecting a brocade sample image, extracting the structural features of a brocade sample pattern, and building a multi-mode joint embedding space; analyzing a design intention input by a user into a multi-modal condition vector, and mapping the multi-modal condition vector to a multi-modal joint embedding space to obtain condition embedding features; introducing a structure perception attention mechanism into the noise prediction network of the diffusion model, generating a structure constraint feature map, and in the reverse denoising process of the diffusion model, carrying out hierarchical fusion on the condition embedding features and the structure constraint feature map to obtain a brocade generation result; and performing brocade process feasibility verification on the brocade generation result, and outputting a brocade pattern. According to the method, the intelligent level and the production efficiency of brocade pattern generation can be improved by combining a structure perception attention mechanism and a hierarchical fusion strategy.

This application discloses a method for identifying and classifying furniture surface defects based on image detection. The method includes: acquiring an original image of the furniture surface; converting the original image into luminance and chrominance components to obtain a standard luminance map; generating a candidate region set based on the standard luminance map and chrominance components; calculating the structure tensor based on the standard luminance map using overlapping local windows to determine the local principal direction and directional validity, and constructing a pixel-level principal direction field and directional validity map; further performing directional-guided background reconstruction on the standard luminance map based on the principal direction field and directional validity map to obtain a background reconstruction map and generate a residual response map; finally, extracting texture interpretation features and defect interpretation features for each candidate region, inputting them into a competition suppression model for discrimination, determining the defect category of the retained regions, and outputting the defect location and defect category. This invention is applicable to the identification and classification of furniture surface defects with directional textures.

The invention discloses a refrigerator oil wear resistance detection device, which comprises a pressure experiment cabin, the pressure experiment cabin is connected with one end of a three-way pipeline through a pipeline, and the other two ends of the three-way pipeline are respectively connected with a vacuum system and a coolant system. A vacuum valve, a refrigerant valve and a constant pressure valve are correspondingly arranged on the pipeline between the vacuum system, the coolant system and the pressure experiment bin and the three-way pipeline, and an agent inlet valve is arranged on the pipeline between the pressure experiment bin and the constant pressure valve. The method comprises the following steps: setting the vacuum degree and temperature of a pressure experiment bin, injecting a refrigerant into the pressure experiment bin, driving a ring block to rotate through a main shaft, enabling the ring block to be in contact friction with a test block, and further judging the lubricating performance of the test oil according to the width and depth of a grinding crack of the test block and the weight loss before and after friction. And the method has good discrimination.

The present invention relates to the field of environmental traceability technology, and specifically to an environmental traceability tracking system for a Juncao production process, comprising a data acquisition module, a data analysis module and a traceability tracking module. The data acquisition module is used to collect environmental data and quality data of Juncao. The data analysis module is used to construct an abnormal data analysis model of an isolated forest algorithm optimized by a wolf pack algorithm, and to construct an SVM-LSTM network model optimized by a dung beetle algorithm. The traceability tracking module is used to mark the Juncao production process, and to input abnormal data into the Juncao traceability analysis model, and to trace the source of the abnormal data. The present invention uses an isolated forest algorithm optimized by a wolf pack algorithm to analyze abnormal data, has better distinguishing ability, is more adaptable, and improves the accuracy of anomaly detection. At the same time, it combines the SVM-LSTM network model optimized by the dung beetle algorithm to perform environmental traceability tracking, which speeds up the convergence of the model, makes the training process more efficient, and improves the overall operating efficiency.