8results about How to "Improve classification" patented technology

The invention discloses a multi-dimensional training method and device for a support vector machine, electronic equipment and a computer readable storage medium, and the method comprises the steps: carrying out the discretization of a training sample data set, and obtaining a discretized data set, the discretized data set comprises a plurality of different attributes, and each attribute corresponds to a plurality of feature vectors; calculating a classification contribution parameter of each attribute feature vector to obtain a plurality of classification contribution parameters; performing data mapping on the plurality of classification contribution degrees by using a kernel function to obtain a target function; and optimizing and training the objective function by using a gradient descent algorithm to obtain a support vector machine model. According to the method, different dimension data are mapped through the kernel function, the data gain weight can be determined, the linear separable effect of the mapped dimension data can be improved, and then the classification and analysis capability of the SVM can be improved.

The application discloses an improved infrared ship image semantic segmentation method and system of a domain adaptive Deeplab model, and belongs to the field of visual semantic segmentation. s and corresponding labels Y S ; an infrared ship image dataset X t is obtained, and the infrared ship image dataset X t is divided into a training set X train and a test set X test ; an improved domain adaptive Deeplab model is constructed; the improved domain adaptive Deeplab model is subjected to semantic segmentation by using the visible light ship image dataset X s , the labels Y S and the training set X train , and an image semantic segmentation result is obtained. The average intersection over union of the infrared ship image dataset is improved by modifying the model architecture and the training strategy of the Deeplab, the accuracy of the trained deep convolutional neural network model is higher, and therefore the classification and positioning capability of the network model for ships in infrared images is improved.

The invention discloses a cultivated land non-agrochemical intelligent monitoring method and system based on multi-temporal SAR and optical image fusion, and belongs to the technical field of remote sensing monitoring and land resource management. The method solves the problems of information loss and misjudgment caused by cloud and mist shielding or noise interference of a single remote sensing data source. According to the scheme, the method comprises the following steps: acquiring and preprocessing a multi-temporal SAR and an optical image sequence; carrying out space registration; constructing a double-branch deep neural network, respectively extracting radar scattering features and spectral texture features, and fusing the radar scattering features and the spectral texture features by adopting a cross attention mechanism; generating a change intensity graph based on a time sequence change detection module; and extracting non-agrochemical suspected pattern spots through threshold segmentation, and aggregating and outputting a final monitoring result. The method is mainly used for realizing the automatic and high-precision monitoring of the non-agrochemical change of the cultivated land.

The application provides a data table sorting display method and device, equipment and a storage medium. A plurality of data tables, user attributes, user operation data and marking results of different data table importance degrees of a plurality of users are obtained. A preset first number of business hot words matched with the user attributes are obtained from a preset database according to the user attributes of the users. The importance degrees of the data tables in each user group are obtained by accumulating the marking results of each user group, and the users are displayed with the sorted data tables corresponding to the user groups to which the users belong, so that the actual application of each user can be better approached, and more accurate data table sorting can be provided for the users.

The present application relates to the technical field of agricultural product origin identification and traceability, and particularly discloses an agricultural product origin identification and traceability system and method based on characteristic fingerprints, which comprises the following steps: collecting agricultural product samples to obtain a target sample set; marking the target sample set with dominant characteristic fingerprints according to the dominant characteristics of the agricultural products, and performing one-class partitioning on the target sample set according to the dominant marking result to obtain a plurality of one-class sample clusters; marking the target sample set with recessive characteristic fingerprints according to the recessive characteristics of the agricultural products, and performing two-class partitioning on the target sample set according to the recessive marking result to obtain a plurality of two-class sample clusters; comparing the one-class partitioning result and the two-class partitioning result to determine whether to start a successive verification mode or a separate verification mode; based on the successive verification mode, obtaining each verification result, and determining whether to output an actual origin traceability result according to the verification results.

The embodiment of the specification discloses a kind of for nasopharyngoscope image classification method, comprising: obtaining available image;By image classification model, the target feature map corresponding to available image is obtained, to determine the image classification result of available image;Target feature map includes: carrying out multiple rounds of feature extraction operation;First round of feature extraction operation includes: after available image is subjected to one or more convolution processing, feature map is obtained, the feature extraction result of this round of feature extraction operation is obtained by one or more parallel convolution processing to the feature map obtained;Each non-first round of feature extraction operation includes: the feature extraction result of previous round of feature extraction operation is obtained after one or more convolution processing, feature map is obtained, the feature extraction result of this round of feature extraction operation is obtained by one or more parallel convolution processing to the feature map obtained;Parallel convolution processing is used to extract different range of features;The feature extraction result of last round of feature extraction operation is used as the target feature map corresponding to available image.

The application relates to the field of agricultural production and environment monitoring technology, and particularly discloses a soil chaotic data identification method based on multi-agent decision, which comprises the following steps: acquiring sample data of soil and constructing a data set; constructing a multi-agent expert pool and a joint representation network, and training the same by using the data set to obtain a soil heavy metal prediction model; wherein the multi-agent expert pool comprises agents of multiple algorithms; the joint representation network comprises a representation encoder, an expert recognizer and a classification discriminator; collecting observable variable data of soil and inputting the same into the soil heavy metal prediction model to obtain a heavy metal concentration prediction value, a confidence, a data type and resampling suggestions; the multi-agent expert pool and the joint representation network can give point estimation, resampling suggestions, sample-level cognitive divergence, knowledge missing estimation and causal sensitivity and other explanatory indexes, and directly support agricultural management decisions such as whether to resample, whether to manually review and the like.

This invention discloses an automatic detection method for TCT smears, comprising: constructing DeepTCT; iteratively training DeepTCT: searching for cervical cells detected by DeepTCT in each image of the test set; if the classification score of the cervical cell is greater than a set threshold, retaining the label of the cervical cell; otherwise, discarding the label; adding samples that reach the set threshold to the training set, and using them together with the existing images in the training set for DeepTCT training; applying the trained DeepTCT to the test set, and outputting the classification and localization results of cervical cells; repeating the iterative training a set number of times; and applying the final DeepTCT to cervical cell morphology detection. This invention significantly improves the detection performance of DeepTCT for cervical cell morphology features through multiple iterative training, and the method of this invention has higher mAP and mAR compared with existing methods.