6results about How to "Realize classification recognition" patented technology

The invention relates to the technical field of material science and image processing, in particular to an iron-carbon alloy microstructure identification method based on deep learning, which comprises the following steps of: processing and shooting all metallographic samples; the method comprises the following steps of: calculating a gray histogram of a metallographic image, then calculating a cumulative distribution function, then equalizing the image, and finally cutting the preprocessed image to a proper size; inputting the preprocessed picture into a Swin Transform model to carry out deep feature extraction learning on the picture; and inputting the extracted image features of each class of iron-carbon alloy microstructures into a full-connection layer through a LayerNorm layer and an Average Pool layer to train a multi-classifier, thereby realizing classification and identification of the iron-carbon alloy microstructures. According to the method, various microstructures of the iron-carbon alloy can be efficiently classified and recognized, and the accuracy and efficiency of recognizing the microstructures of the iron-carbon alloy are improved.

The application discloses a kind of sonar image classification methods of combined SLIC superpixel and graph attention network.It includes the following steps: according to the imaging principle of two-dimensional forward-looking sonar and side scan sonar respectively and prior information when imaging: the sonar image after correction is based on the image pre-segmentation of improved DeepLabV3+ network, SLIC superpixel algorithm is used to construct Graph (graph) structure data: construct the sonar image classification model based on GAT (graph attention network), the sonar graph structure data built is sent into network to complete the training and test of model;Verify the importance of pixel feature and spatial position feature.The application discloses a kind of sonar image classification methods of combined SLIC superpixel and graph attention network, which fully utilizes the space position relationship of acoustic shadow area, target area and shadow area by SLIC superpixel method and graph attention network, so as to realize higher accuracy of sonar image classification identification of sonar image by combining pixel feature and spatial geometric feature.

The embodiment of the application provides a robot control method and device based on an electroencephalogram signal, a computer readable storage medium and a computer program product, the method comprising: controlling an electroencephalogram acquisition device to collect original electroencephalogram signals in real time, preprocessing the original electroencephalogram signals to obtain target electroencephalogram signals, and the preprocessing at least comprising band-pass filtering; using a multi-band filter bank to perform multi-band decomposition on the target electroencephalogram signals into a plurality of sub-bands of different frequency ranges, and performing frequency domain transformation on each sub-band to obtain multi-band frequency domain features; inputting the multi-band frequency domain features into a pre-trained deep learning-based Transformer network to obtain a target category, the Transformer network modeling the correlation between different sub-bands based on a self-attention mechanism to realize feature fusion and deep feature extraction of the multi-band frequency domain features; and generating and outputting a control instruction based on the target category. The embodiment can improve the accuracy and stability of robot control.

This invention discloses a visual inspection system and method for transformer tank panels in new energy applications, belonging to the field of transformer tank manufacturing and inspection technology. The system includes a flexible steady-state clamping and conveying mechanism, a transient pulse excitation mechanism, a high-frequency multimodal visual acquisition mechanism, and a data processing and spatiotemporal response judgment module. By applying a controlled transient thermal pulse to the area to be inspected and simultaneously acquiring static three-dimensional reference surfaces and dynamic three-dimensional response surface sequences, spatial registration and differential processing are performed to obtain a transient out-of-plane displacement field. Welding defects are then identified based on preset spatiotemporal continuity anomaly judgment conditions, and the system combines two-dimensional surface texture images to classify and identify internal latent defects and surface penetration defects. This invention is suitable for online inspection of transformer tank panels.

The present invention belongs to the technical field of household appliance identification, and discloses a non-invasive electrical appliance identification method, system, medium, equipment and terminal. The current of the electrical appliance is synchronously sampled by using a current transformer and a voltage zero-crossing detection circuit; the synchronously sampled current data is processed by using an MCU to obtain the parameters of the fundamental phase of the current, the amplitude ratio of each harmonic to the fundamental wave, and the energy ratio, and the characteristic quantity of each electrical appliance is calculated by using the current parameters; the type of the electrical appliance that has changed is determined by extracting dynamic characteristic values ​​from the changing data, and the type and quantity of the current electrical appliance are intelligently identified by combining accumulation and CUSUM. The present invention takes into account low cost, low power consumption, and high recognition accuracy, and is easy to industrialize and promote; it has a learning function, can autonomously learn new electrical appliances, and the identification of electrical appliances is not limited to the types of electrical appliances built-in at the time of leaving the factory; training a small amount of data for a single electrical appliance can accurately identify the combined electrical appliances.

The invention provides a self-powered building vibration sensing system based on a triboelectric nano-generator and a synaptic transistor. The self-powered building vibration sensing system comprises the triboelectric nano-generator, the organic field effect synaptic transistor and an LED lamp. The triboelectric nano-generator is of a sandwich structure and sequentially comprises a friction layer, an electrode and a packaging layer from top to bottom. The organic field effect synapse transistor is composed of a substrate with an insulating layer, an active layer and a top electrode from bottom to top. The LED lamp is connected with the source electrode and the drain electrode of the top electrode of the organic field effect synapse transistor; detecting a vibration signal generated by the building structure by using the triboelectric nano generator, and converting the vibration signal into an electric signal; a synapse transistor is used for processing the electric signal, simulating the function of a biological synapse, and processing and evaluating the vibration signal; the LED lamp generates a light signal when the output response of the synaptic device exceeds a preset threshold.