8 results about "Deep cnn" patented technology

A Large Language Model (LLM) is configured to automatically label non-labeled textual data-items for the purpose of creating a training dataset for training a Machine Learning (ML) model. The ML model is thus trained on LLM-labeled textual data-items; and the ML model can be deployed to classify new or incoming documents or messages or other textual data-items. Additionally, a Vision and Language Model (VLM) or a Large Multimodal Model (LMM) or a large multiple-modalities model (LMM) can process data from two or more modalities (visual data, textual data), and is configured to automatically label non-labeled images for the purpose of creating a training dataset for training a Deep Neural Network (DNN) or a Deep Convolutional Neural Network (Deep CNN) model. The DNN model is thus trained on VLM-labeled images; and the DNN model can be deployed to classify new or incoming images.

The application relates to a model pruning method of an interpretable CNN classification model, belongs to the field of image compression, and solves the problems of high operation complexity, large time and memory consumption and difficulty in deployment on terminal equipment of an existing deep CNN model, and solves the problem of lack of interpretability of an existing model pruning algorithm. The method comprises the following steps: inputting a training picture into a neural network model to be pruned, and extracting a feature map matrix of each convolution layer; upsampling the feature map matrix to the size of the input picture, and then performing a normalization operation to construct a saliency map; multiplying the saliency map and the input picture element by element to construct a weighted input picture; subtracting the input picture from the weighted picture element by element to construct an attention region occlusion map; inputting the attention occlusion map into the model to be pruned, observing the change of model accuracy as an importance score of the channel, and pruning the channel to obtain a pruned lightweight model. The application realizes high pruning rate of the model and improves the interpretability of the pruning process.

The application discloses a pumping unit noise positioning method based on deep learning, and steps are as follows: S1, local seismic trace sets of several fixed sizes are obtained by using sliding window single-step sliding to block seismic data by columns, if the number of seismic traces containing pumping unit noise reaches threshold Th1, it is marked as 1, and if the number of seismic traces containing pumping unit noise is lower than threshold Th2, it is marked as 0; S2, the local seismic trace sets are pretreated, the energy spectrum is calculated, mean filtering is carried out, then down-sampling is carried out, and part of data is randomly selected as a training set, and the remaining part is used as a test set; S3, a deep CNN network is built, and the training set obtained in step S2 is used to train the CNN network, and in the training process, the misclassified data is used to supplement the training set for repeated training; S4, the test sample is used to test the CNN network trained in step S3, and the positioning function is quantitatively evaluated; S5, the width of the positioned pumping unit noise is estimated according to the positioning result in step S4.

The invention relates to the technical field of offshore culture intelligent monitoring, and discloses a multi-mode and deep CNN offshore culture intelligent edge detection method and system, and the method comprises the steps: obtaining local multi-depth water sample nutritive salt concentration data and nitrogen and phosphorus isotope composition data of each farm, and generating a local pollution source isotope fingerprint vector; adding Laplacian noise to the isotope fingerprint vector of each farm based on a differential privacy algorithm, and calculating a disturbance fingerprint vector meeting epsilon-differential privacy; combining the combined fingerprint feature space with a hydrological transmission model, and decomposing contribution proportions of different sources in nutritive salt at each observation point by using a federal Bayesian mixture model; and outputting a real-time contribution ratio and a confidence interval of each pollution source based on the trained pollution source identification model. According to the invention, accurate pollution source identification and contribution proportion quantification are realized on the premise of protecting the privacy of each party.

The application discloses a LIBS multi-distance mixed spectrum classification method based on a deep CNN and sample weight optimization, and relates to the technical field of spectrum analysis. The application constructs a deep convolutional neural network (CNN) model and uses the model for classification of laser-induced breakdown spectrum (LIBS) data, and is suitable for analyzing mixed LIBS spectrum data collected at multiple different distances. The core innovation of the application is that different sample weights are given to spectrum samples at different distances in the CNN training process. In a conventional model training method, equal weights are given to all training set spectrum samples, while in the application, the weight of each spectrum sample is specially designed according to the absolute distance value of the training set spectrum sample and the distance difference between the training set spectrum sample and the test set spectrum sample, so that the weights of spectrum samples at different distances are optimized, and thus the classification effect of the deep CNN model is improved. The application has the advantages of no distance correction, efficient training and high accuracy, and can effectively classify LIBS multi-distance mixed spectrum.

A Large Language Model (LLM) is configured to automatically label non-labeled textual data- items for the purpose of creating a training dataset for training a Machine Learning (ML) model. The ML model is thus trained on LLM-labeled textual data-items; and the ML model can be deployed to classify new or incoming documents or messages or other textual data- items. Additionally, a Vision and Language Model (VLM) or a Large Multimodal Model (LMM) or a large multiple-modalities model (LMM) can process data from two or more modalities (visual data, textual data), and is configured to automatically label non-labeled images for the purpose of creating a training dataset for training a Deep Neural Network (DNN) or a Deep Convolutional Neural Network (Deep CNN) model. The DNN model is thus trained on VLM-labeled images; and the DNN model can be deployed to classify new or incoming images.

The invention discloses a gas remote sensing image feature extraction method based on a diffusion model, and relates to the technical field of remote sensing image processing, and the method specifically comprises the steps: obtaining a gas remote sensing image data set; preprocessing the gas remote sensing image data set; constructing a basic CNN network model of the fusion diffusion model; inputting the preprocessed gas remote sensing image into a CNN network model, and obtaining a shallow feature map through a CNN shallow network; inputting the shallow feature map into a pre-trained diffusion module, and carrying out denoising and enhancement processing; and inputting the enhanced feature map into a CNN deep network to obtain a final gas remote sensing image feature. According to the method, the diffusion model is used as an independent module to be embedded between the shallow layer and the deep layer of the CNN network, the diffusion model is used for denoising the feature map extracted by the CNN network, details are enhanced, the problems of noise interference and difficulty in capturing weak signals in the remote sensing image are solved, related feature data and analysis conclusions of the gas remote sensing image are output, and powerful support is provided for subsequent analysis.

A method for intelligent extraction of damage features and construction of health indicators from multi-source data in aircraft structural monitoring is proposed. First, raw monitoring signal data from multiple sources, including acoustic emission waveform stream data and guided wave data, are acquired. A convolutional neural network (CNN) is then constructed to automatically learn deep features from the raw data. Expert knowledge features are extracted from the raw data and fused with the deep features of the CNN. The fused feature vector is then input into a classifier for classification. Next, a feature vector sequence is extracted using sliding window sampling to construct a full-lifecycle feature dataset. A health indicator construction model based on a deep CNN is established, employing a contrastive learning loss function as the optimization objective and iteratively training the model using gradient descent. Finally, the health indicators from the acoustic emission waveform stream data and guided wave data are fused from multiple sources. By utilizing the consistent changes in the data at the time of damage, the damage evolution process is comprehensively judged, and a fused health indicator is output. This invention improves the accuracy and reliability of the health indicators.