13 results about "Small data sets" patented technology

The application belongs to the technical field of unmanned aerial vehicle image detection, and in particular to an unmanned aerial vehicle image detection method based on an improved RT-DETR model. The method comprises the following steps: selecting a public unmanned aerial vehicle image dataset, dividing the training set, constructing a small dataset, training and evaluating the improved RT-DETR model on the small dataset, and then implementing the model on the complete dataset to obtain the final data. The application enhances the information of small targets in the fusion process without increasing the number of model parameters. While introducing S2, a new fusion module mainly composed of DSConv is used to replace the fusion module of the benchmark model, which balances the parameter quantity of the model while improving the small target detection accuracy. Compared with the previous algorithm, the improved model has higher accuracy and smaller parameter quantity, and can provide protection for accurate detection of unmanned aerial vehicle images.

The application relates to the technical field of archive construction, and provides a text attribute relation extraction method based on a small data set pre-training model, a pre-training model based on an ERNIE structure is publicly fine-tuned to a relation extraction model in a business background through a public pre-training model UIE which can be started by fine-tuning only a small amount of labeled data, and on the basis of the model, Before_pipeline data processing before prediction and After_pipiline data processing after prediction are constructed in combination with business text features to improve result reliability, the problems of lacking a labeled data set, UIE violent sentence splitting leading to errors, and low extraction result accuracy of the UIE are solved, and the problem that an existing traditional relation extraction model depends on a large amount of high-quality data sets manually labeled is solved.

Methods and systems for ultrasound imaging are provided. In one example, a method includes receiving, with a wireless handheld probe assembly, ultrasound signals of a region of interest; generating, within the wireless handheld probe assembly, a plurality of received digital signals based on the received ultrasound signals; generating each of a larger data set and a smaller data set from the plurality of received digital signals; transmitting the smaller data set from the wireless handheld probe assembly to a hub via a lower bandwidth wireless connection; transmitting the larger data set from the wireless handheld probe assembly to the hub via a higher bandwidth wireless connection; generating, at the hub, a low resolution image from the smaller data set and a high resolution image from the larger data set; and transmitting the low resolution image from the hub to a first display and the high resolution image from the hub to an electronic device.

The application discloses a kind of multi-dimensional dynamic model under the evaluation method of soil salinization to cultivated land productivity, it is related to the field technical field of agricultural resources and environment, and the multi-source data set of fusing remote sensing, ground monitoring, meteorological and farmer management data is first constructed in the scheme, and the minimum data set of soil physical, chemical, biological properties is screened after fusion preprocessing;Soil quality index (SQI) and cultivated land productivity index (CPI) are calculated again;Subsequently, the multi-dimensional dynamic evaluation model containing key salinization index screening, effect quantification, mechanism model and crop sub-model is constructed;Finally, the productivity influence dynamic update is realized in combination with real-time data of Internet of Things, and the optimal improvement technology mode is simulated and screened by structural equation model simulation management scene. The method is comprehensive in data coverage, accurate in evaluation, dynamic in model and targeted, can reveal the influence mechanism of salinization on cultivated land productivity, provide practical decision support, and help efficient and sustainable use of saline-alkali soil.

This invention relates to a remote sensing extraction method for pear orchard areas based on the Re-UNet model, which overcomes the shortcomings of inaccurate classification results and low efficiency in pear orchard area extraction from remote sensing images compared with existing technologies. The invention includes the following steps: acquiring a remote sensing image dataset; constructing the Re-UNet pear orchard area extraction model; training the Re-UNet pear orchard area extraction model; acquiring and preprocessing the remote sensing images of the pear orchard areas to be segmented; and obtaining the remote sensing extraction results for the pear orchard areas. Based on the UNet semantic segmentation model, this invention solves the overfitting problem that easily occurs in small datasets. It also incorporates spatial and channel attention mechanisms and a residual module, further enhancing the feature transfer and cumulative integration characteristics of pear orchard areas in high-resolution remote sensing images, effectively reducing the "salt and pepper" phenomenon and misclassification, and improving the overall segmentation accuracy.

The application discloses a small data set craniomaxillofacial translation method based on GAN, which comprises the following steps: 1, collecting skull and facial CT image data; 2, performing image preprocessing, three-dimensional reconstruction and fairing treatment on the skull and facial CT image data to obtain complete three-dimensional models of the skull and the face; 3, placing the three-dimensional models of the skull and the face in the Frankfurt coordinate system to perform normalization operation; 4, performing vertical mapping of the three-dimensional models of the skull and the face on the XOZ plane in the Frankfurt coordinate system to obtain the front view images of the skull and the face; 5, introducing a Gaussian pyramid into a GAN network to construct a network model PCC-GAN for skull and facial translation; 6, training network parameters of the pyramid cycle consistency generative adversarial network model PCC-GAN; and 7, placing the skull and facial images into the craniomaxillofacial translation model PCC-GAN to generate two-dimensional skull and facial images, and more accurate and real facial images can be generated under the condition of less point cloud data.

The application discloses a small-dataset parallel training system based on GPU, adopts a hierarchical modular architecture design, builds a complete training pipeline from data input to result output, and comprises seven modules, namely, a data input layer, a data preprocessing module, a GPU batch processing module, a memory management module, a parallel processing module, a model training module and an output layer, wherein each module independently realizes exclusive functions and cooperates with each other, forms a single-GPU optimized training system suitable for small datasets, solves the problem that differences between small-dataset samples are large and individual differences are significant, and realizes efficient and stable training of small datasets in a single-GPU environment.

The application discloses a power-off holding method for a communication device, and particularly relates to the technical field of communication data protection, and comprises the following steps: after detecting that the main power supply is powered off, the maximum instantaneous current of the backup power supply is acquired, the key data is screened according to the current capacity, and a minimum data set is formed, address verification and pre-charging preparation are completed after being divided into multiple atomic write units; during the backup power supply power supply, write operations are sequentially executed, when the backup power supply output voltage approaches a preset voltage safety threshold, the remaining write is terminated, and the state mark and integrity check information of the completed unit are written into an independent storage area; the application controls the write by acquiring the maximum instantaneous current of the backup power supply, guarantees the write start safety, improves the write reliability and fault recovery capacity through data screening and atomic write unit construction, controls the write termination time through voltage dynamic monitoring, prevents data abnormalities, and enhances the stability and data integrity of the communication device power-off holding.

The application belongs to the field of organic polymers, and particularly discloses a method for predicting the performance of grafted polypropylene for high-voltage power cables. In the method, a performance prediction model of the grafted polypropylene is constructed based on a multilayer perception model and a random forest model, which solves the problem that traditional machine learning models mainly rely on a large amount of data and are not suitable for small data sets. The relationship between the graft structure and the mechanical performance is quantitatively established from experimental data, the required data amount is small, the data is easy to obtain, and the prediction accuracy is high, thereby providing a reliable graft group for the grafted polypropylene material for high-voltage power cables.

This invention discloses a deep learning-based environmental detection method and system, relating to the field of gas detection technology. It involves real-time acquisition of the raw response signals of a gas sensor array, followed by signal segmentation and context-aware reconstruction to obtain an enhanced sub-signal set. The target enhanced sub-signals are then transformed into a spatiotemporal matrix to obtain two-dimensional features. Temporal features and local temporal features are extracted along the time channel dimension to obtain temporal enhancement features. Sensing features are extracted along the sensor channel dimension, and inter-sensor correlation features are extracted to obtain sensing enhancement features. All sensing enhancement features are flattened and input into a fully connected layer, then passed through a Softmax layer to obtain the gas category recognition result. This method addresses the problems of insufficient cross-channel feature fusion and inadequate feature representation in existing methods through multi-dimensional feature fusion and attention mechanisms. It also expands the training samples from the original signals through signal augmentation and reconstruction, adapting to small dataset scenarios, thereby improving the accuracy of gas recognition.

A system develops classifiers that can aid medical professionals by diagnosing whether or not a patient has a medical condition like pneumothorax. The system breaks the task into multiple steps, using YOLOv4 to extract relevant regions of the video and a 3D sparse coding model to represent video features. Given the difficulty in acquiring positive training videos, the inventors trained a small-data classifier with a maximum of 15 positive and 32 negative examples. To counteract this limitation, the inventors leveraged subject matter expert (SME) knowledge to limit the hypothesis space, thus reducing the cost of data collection. The inventors present results using two lung ultrasound datasets and demonstrate that the inventors' model is capable of achieving performance on par with SMEs in pneumothorax identification.

The application discloses a single budget multi-round federated learning incentive mechanism method, device, equipment and medium, and the method comprises the following steps: receiving the total budget of a task publisher, a learning task and a standard small data set; according to the standard small data set, the quality of the local model of all task participants in the round is evaluated; according to a selection strategy and a payment function, in combination with the quality evaluation value of the local model of each task participant in the round, the task participants participating in the federated learning in the round are selected, and the remuneration of the task participants is determined; if the total budget is greater than 0 after the payment of remuneration, the global model is sent to the selected task participants, so that the local model of each task participant is locally trained; the local model trained by each task participant is received, and the global model is aggregated by using an aggregation method. The application designs a feasible incentive mechanism for non-independent and identically distributed federated learning, and can further promote the landing use of federated learning application.

The application provides a gesture recognition method, system and device based on surface electromyogram signals, the method comprising: initializing a model in a server, establishing a joint model; a client collects local data; the server broadcasts the joint model to the client; the client trains the joint model using its local data on the client side, forming a client model; the parameter matrix of the client model is uploaded to the server; the server obtains the parameter matrix of a new joint model based on the parameter matrix; after reaching a preset update round, a final joint model is obtained. This scheme can effectively reduce the influence of cross-domain under the condition of data scarcity, combine multiple clients with small data sets, train a joint model with strong generalization ability under the premise of protecting data privacy, and when new data is encountered, parameter fine-tuning is performed on the joint model, so that a network model with good performance on new data can be obtained in a short time.