6 results about "Gabor transform" patented technology

The application discloses a kind of safety rope operation state monitoring method and system based on double-scale differential entropy feature, it is related to safety rope monitoring technical field.Method includes collecting original stress time series data and marking, obtains stress data set with time series marking;High-frequency impact component of original stress time series data with time series marking is extracted using dynamic narrow window Gabor transform, low-frequency entropy value is calculated using fixed wide window integration combined with differential entropy operator, high-frequency impact component and low-frequency entropy value are spliced, and double-scale differential entropy feature is constructed;Neural network model is constructed and trained by double-scale differential entropy feature;The trained neural network model is deployed in safety rope monitoring terminal to carry out aerial work safety rope real-time monitoring.The application significantly improves the accuracy and robustness of aerial work safety rope state identification.

The invention discloses a pre-stack seismic reflection feature extraction method and device based on two-dimensional Gabor transformation. The method comprises the steps of obtaining a two-dimensional image of pre-stack seismic data; performing convolution processing on the two-dimensional image by using two-dimensional Gabor kernels of different directions and different scales to obtain a plurality of Gabor features corresponding to different directions and different scales, the Gabor features being complex number domain values; calculating the amplitude of each Gabor feature, wherein the amplitude of each Gabor feature is used for representing the reflection feature in the corresponding direction and scale; and combining the amplitudes of all the Gabor characteristics to form a Gabor characteristic matrix so as to represent the integral reflection characteristics of the pre-stack seismic data. According to the method, the stability of seismic wave reflection feature extraction and the reliability of results can be improved, and geological abnormal information identification is realized.

The application discloses a geological radar echo multi-scale analytical processing method based on Gabor transformation, and utilizes adjustment of Gabor filter parameters to obtain feature enhancement images of different targets under different scales, matches the feature enhancement images with disease maps to identify targets in ground penetrating radar echo data. The application effectively solves the technical problem that the existing echo analytical technology of ground penetrating radar detection has high difficulty in identifying different disease targets and high subjective factors, filters radar echoes by adjusting different bandwidths, different amplitude type ratios, different wavelengths, different directions and other parameters, obtains a data enhancement method under different parameters, realizes identification of different feature targets, and effectively extracts features of different disease targets under different scales.

This invention discloses a robust zero-watermarking method and system for medical images based on local minimum adaptive two-dimensional empirical wavelet transform (2DEWT), polar harmonic transform (PHT), and Gabor transform. The method includes the following steps: Step 1, copyright watermark encryption process; Step 2, copyright watermark embedding process; Step 3, copyright watermark extraction process; Step 4, copyright watermark decryption process; Step 5, copyright system usage process. This invention uses 2DEWT in the extraction of medical image features and, based on the characteristics of medical images, designs an adaptive improvement strategy that directly uses the local minimum points of the frequency domain energy curve as sub-band boundaries. The local minimum points are essentially the boundary points of different feature components in the frequency domain of medical images: the boundaries between low-frequency global structures and high-frequency lesion details, and the boundaries of lesion textures in different directions (such as blood vessel orientation and lesion edge angles) all exist in the form of local minimum points. Two-dimensional empirical wavelet transform is performed on medical images to extract low-frequency subbands. Gabor transform is applied to the low-frequency subbands, and PHT is applied to the high-frequency subbands. Image features are obtained through feature fusion, and the copyright watermark is embedded by obtaining the key through XOR with the copyright watermark.

The invention discloses a cross-domain radiation source individual identification method and system based on a weak supervision data enhancement network, belongs to the technical field of wireless communication, and solves the problems that the identification accuracy is reduced and the model robustness is insufficient under a cross-receiver condition. The method comprises the following steps: collecting an original wireless signal, and carrying out preprocessing; analyzing the preprocessed signal by adopting short-time Fourier transform and Gabor transform to obtain a time-frequency image; performing multi-scale feature extraction on the time-frequency image by using an Inception hybrid network to obtain a feature map; generating an attention map by adopting a weak supervision attention learning technology, and obtaining a sensitive area according to the attention map; adopting bilinear attention pooling to aggregate the sensitive area and the feature map, and improving the feature expression ability; and based on a data enhancement technology of attention guidance, performing cutting and discarding operation on the attention map, generating enhanced features, performing multi-classification identification on the enhanced features, and outputting a radiation source category. The method is suitable for individualized scene distinguishing of the transmitting equipment.

The application discloses a rotating machinery modulation feature extraction method based on prior traversal envelope spectrum, comprising the following steps: (1) collecting and calculating the spectral coherence function of the rotating machinery monitoring signal based on Gabor transform; (2) performing a slicing operation on the spectral coherence function along the carrier frequency direction, and calculating the harmonic intensity vector of different harmonic cluster structures; (3) calculating the harmonic intensity matrix containing all carrier frequencies; (4) when the prior information is a single frequency, a single weighting function is calculated; when the prior information is a certain frequency range, a composite weighting function is calculated; (5) performing a threshold filtering operation to obtain a prior traversal weighting function; (6) performing prior traversal weighting processing on the spectral coherence function to obtain prior traversal spectral coherence, and further obtaining prior traversal envelope spectrum. By using the application, interference can be adaptively filtered and signal features can be extracted and enhanced, and good extraction effect can be realized even under strong noise interference.