9 results about "Detection performance" patented technology

The invention discloses a steel bar detection method based on a deep convolutional neural network. The method comprises the following steps: 1) data acquisition; 2) data processing; 3) constructing and training of a model; 4) model evaluation; 5) model deployment. The target detection algorithm based on the deep convolutional neural network is applied to reinforcing steel bar detection and counting, and a multi-scale and deformation-tolerant reinforcing steel bar detection network framework is provided. Cascade R-CNN with good detection performance, a feature pyramid network capable of effectively solving the multi-scale detection problem, a more stable group normalization module, and a deformable convolution module capable of improving the spatial geometric deformation capacity of network learning and other modules are integrated in the frame. Compared with a traditional steel bar detection method, the network framework is higher in detection precision and better in generalization, and can be deployed in a remote server or a mobile device.

The invention provides a radio frequency interference suppression method for a high frequency surface wave radar based on tensor subspace projection. Depending on the distance correlation and direction characteristics of radio frequency interference, joint estimation is performed on a radio frequency interference subspace from multiple dimensions by tensor decomposition. The implementation processcomprises the following steps: performing Doppler sliding window processing based on radar data to finish data segmentation; performing radio frequency interference detection; judging whether interference frequency points exist in a currently-selected data block or not; if radio frequency interference exists, constructing a third-order tensor by using long-distance metadata without sea echo of multiple antenna channels to serve as a training tensor; performing tensor decomposition on the training tensor by using high-order singular value decomposition; estimating a radio frequency interference subspace; and performing distance sliding window and subspace projection processing to obtain data without radio frequency interference. The method can suppress radio frequency interference to a great extent, improve data quality and retain a target signal completely, has a good application effect, improves the detection performance of a radar system, and has important practical application value.

The invention discloses a blood pressure measurement method and system. The method comprises the following steps that a blood pressure measurement expression with correction factors is set according to the relationship between shear stress and the blood flow velocity V; characteristic parameters of the blood flow velocity V of a measured object are obtained, meanwhile the blood pressure reference value of the measured object is directly collected through a pressure gauge, and inverse calculation is conducted on the blood pressure measurement expression to obtain a correction factor value; when the practical blood pressure measurement is conducted, the correction factor value of the corresponding measured object is read from a database, and the blood pressure measuring result is calculated on the basis of the blood pressure measurement expression and the characteristic parameters, collected in real time, of the blood flow velocity. The blood pressure measurement method and system are simple, safe, convenient in test, and can achieve remote real-time monitoring of the blood pressure of the user, are good in real-time detection performance, can cope with various accidents of sudden illness of a blood-pressure-disease patient timely, eliminate the damage to the human body by a traditional direct measurement method, store the correction factors and characteristic constant values of the different measured objects for blood pressure calculation, and are high in measurement accuracy.

The invention discloses a method for preparing a graphene chemically modified electrode through in-situ growth, which mainly comprises the following steps: (A) substrate cleaning; (B) in-situ growth of graphene: placing the substrate obtained by the step (A) in a clean quartz boat, and placing the quartz boat in a quartz tube of a horizontal resistance furnace; introducing argon to remove oxygen in the quartz tube; under the protection of the argon, heating the quartz tube to the growth temperature of graphene; preserving heat for 1-180 minutes while introducing carbon source gas and hydrogen instead; then, cooling to room temperature under the protection of argon, and taking out a product to obtain a substrate with in-situ grown graphene; (C) connecting the substrate with graphene of in-situ growth, obtained by the step (B), with a wire or directly putting the substrate into a clamp to obtain the graphene chemically modified electrode. The preparation process of the method is simple, the preparation efficiency is high, and the method is suitable for large-scale production; the prepared graphene chemically modified electrode has good quality and excellent detection performance.

The invention relates to the field of anomaly detection and new high-risk alarm identification by adopting a One Class SVM (Support Vector Machine), and particularly provides an anomaly alarm identification method and device based on incremental learning, a server and a storage medium, the method comprises the following steps: firstly, training a One Class SVM model by utilizing an impure original data set to identify an anomaly alarm; and then, enabling a safety operator to regularly mark abnormal alarms obtained by the model, and iteratively optimize the model based on incremental learning, so that the model detection performance is improved, and missing report and false report are reduced.

The invention provides an anti-large frequency offset solution method based on a 5G random access preamble long sequence, and the method employs the cyclic prefix of a random preamble to calculate and correct the decimal frequency offset, and solves the influence of a secondary peak generated by the decimal frequency offset on the detection performance of a preamble sequence. After fast Fourier transform, five paths of frequency domain data with different integer frequency offsets are constructed in a mode of extracting effective subcarrier frequency domain data from the subcarrier frequency domain data; then, matching is carried out by adopting different root sequences according to the characteristic of preamble cyclic shift, and the maximum value is judged through search window shift, so the processing time of sequence detection and the hardware resource consumption are reduced; and the influence of integer frequency offset on random access caused by inaccurate frequency offset calculation and application scenarios under the condition of low signal-to-noise ratio is solved by utilizing a five-path parallel matching scheme.