110 results about "Local binary patterns" patented technology

The invention relates to the technical field of remote sensing data processing and glacier area calculation, in particular to an unmanned aerial vehicle and satellite remote sensing data fused glacier area calculation method, which comprises the following steps of: cooperatively acquiring a satellite multispectral image and unmanned aerial vehicle high-resolution optical and LiDAR data, performing time synchronization, high-precision space registration and data enhancement processing, and calculating the glacier area through the unmanned aerial vehicle and satellite remote sensing data fusion. A satellite image glacier macroscopic feature and an initial mask are extracted by using a convolutional neural network and an NDSI / NDWI algorithm, and unmanned aerial vehicle image microscopic texture, edge and topographic features are acquired through a local binary pattern, edge detection and LiDAR point cloud; based on pyramid layering and a conditional random field, adopting a variance weighting algorithm to realize multi-scale feature level fusion; and after segmentation through an Otsu algorithm, calculating the area through a pixel counting method and introducing gradient correction, and evaluating the reliability through three types of precision. The method breaks through the limitation of a single data source, fuses macroscopic and microscopic features, improves the boundary positioning precision and calculation efficiency, and is suitable for glacier dynamic monitoring in a complex environment.

The invention discloses an image multi-modal feature extraction method, a ground feature classification and recognition method and a GIS image generation method. Comprising the following steps: firstly, extracting texture features from a target image by adopting a multi-directional statistical method fusing rotation invariant coding of a local binary pattern and a gray-level co-occurrence matrix, and extracting color features from the target image by adopting an LAB-HSV dual-color space collaborative analysis method to obtain features of different modes of the target image; and then according to the information values of the extracted color features and texture features, adjusting the weight ratio corresponding to the color features and the texture features so as to optimize the recognition precision of the classification model on complex ground features. And finally, according to a weight ratio corresponding to the color feature and the texture feature, performing weighted fusion on the color feature and the texture feature to obtain a corresponding multi-modal feature vector.

The invention discloses an intelligent identification system for an osteoporosis area, belongs to the field of image processing calculation, and aims to solve the problems of limited technical coverage and lack of a dynamic optimization mechanism. In a feature extraction stage, a system combines traditional image processing and deep learning technologies in parallel, extracts bone trabecula multidirectional texture features by using a Gabor filter and a local binary pattern algorithm, analyzes bone contour curvature by combining Sobel edge detection and morphological operation, constructs geometric morphological parameters, and performs feature extraction on the bone trabecula. The local branch focuses on the porosity and arrangement rule of the bone trabecula by adopting a 3D convolutional network, the global branch is embedded into a compression excitation module based on an improved MobileNetV3 network to strengthen the overall morphological expression of the bone, the response intensity of a local microstructure is enhanced by space attention, the weight of global bone topological characteristics is calibrated by channel attention, and a multi-scale splicing strategy is combined, so that the overall morphological expression of the bone trabecula is optimized. And finally, outputting a fusion feature matrix after noise suppression, and remarkably improving the expression ability of pathological features.

The invention discloses an Internet of Things malicious traffic detection method and device and a storage medium, and the method comprises the following steps: S1, collecting malicious traffic data of Internet of Things equipment, and classifying and constructing a malicious traffic sample library; s2, converting the malicious traffic data into a grayscale image by adopting a text imaging method, and performing normalization, noise reduction and feature enhancement; s3, extracting local binary pattern features, global image features and high-order local automatic correction features, and constructing a malicious traffic feature database; s4, training the malicious traffic detection model by adopting a space-time capsule Transform, extracting space and time features, and performing classification training; s5, collecting real-time traffic data, calculating an abnormal score by using the trained model, and judging a traffic category; and S6, sending a detection result to a command and control server, triggering an alarm, and sending a blocking or isolating instruction to the attacked device. The method improves the accuracy and real-time performance of malicious traffic detection, and is suitable for security monitoring and protection of the Internet of Things.

The invention relates to the technical field of image processing, and discloses a morphology-based image texture feature extraction method, which comprises the following steps: an image preprocessing step: carrying out graying processing on an original image, and carrying out smoothing processing on a grayscale image by using a Gaussian filter; a morphological operation step: selecting structural elements suitable for image texture characteristics, performing expansion and erosion operation on the smoothed image, performing opening operation on the expanded image, and performing closing operation on the eroded image; and an adaptive feature selection step: carrying out local binary pattern feature extraction on the image after morphological processing, adaptively determining a threshold according to statistical distribution of LBP features, and segmenting the image into a foreground and a background. The image texture feature extraction method based on morphology aims at solving the problems that feature extraction is inaccurate, noise interference is likely to happen, and feature combination is complex in an existing image texture feature extraction method.

The invention provides a zinc-plated aluminum-magnesium steel strip surface coating image enhancement method, and relates to the technical field of image enhancement. A multi-scale feature extraction module provided by the invention adopts a method of combining an improved adaptive threshold local binary pattern and a directional gradient feature to extract edge textures and structural features under different scales; the feature selectivity enhancement module adopts an improved multi-scale direction response suppression and sparse activation enhancement method to perform direction saliency selection and detail region response improvement on the multi-scale feature map; and the cross-scale feature fusion module constructs a structural response graph based on local contrast variance, realizes pixel-level weighted fusion of a multi-scale feature graph, and finally outputs an enhanced zinc-aluminum-magnesium-plated steel strip surface coating image.

The invention discloses a thyroid cancer pathological image classification method and system based on multi-modal feature fusion and Bayesian optimization, and relates to the technical field of medical image processing, and the method comprises the steps: collecting a thyroid ultrasound image data set, and carrying out the preprocessing operation; extracting an improved local binary pattern feature, a Haralick texture feature and a VGG16 depth feature, and constructing a mixed feature space; and splicing the features in the mixed feature space into a 4119-dimensional mixed feature vector, and carrying out feature importance screening by utilizing ExtraTres. Improved local binary pattern features, Haralick texture features and VGG16 depth features are fused, a mixed feature space is constructed, feature importance screening and PCA dimension reduction are performed by using ExtraTrees, multi-level features of the image are effectively extracted, the accuracy of benign and malignant thyroid nodule classification is remarkably improved, meanwhile, network hyper-parameters are dynamically adjusted through a Bayesian optimization algorithm, and the classification accuracy of benign and malignant thyroid nodules is improved. And model convergence is accelerated in combination with a cosine annealing strategy, so that the generalization ability of the model is enhanced, and the model can be excellently expressed on different data sets.

The invention provides a bearing fault diagnosis method based on a one-dimensional local binary pattern and a Hankel matrix. The bearing fault diagnosis method comprises the following steps: acquiring a discrete vibration signal; performing first-order differential operation on the discrete vibration signal to obtain a differential signal; performing inherent time scale decomposition on the differential signal to obtain an inherent rotation component signal; performing quantization and signal reconstruction on each inherent rotation component signal by taking a root mean square as a quantization criterion of a one-dimensional local binary mode method to obtain a decimal feature signal; constructing a Hankel matrix of the decimal characteristic signal and performing signal reconstruction according to a covariance matrix of the Hankel matrix; performing spectral analysis on the reconstructed signal, calculating the fault characteristic frequency of the bearing, and then judging the state and the fault type of the bearing through a frequency component obtained through spectral analysis and the fault characteristic frequency of the bearing obtained through calculation. According to the bearing fault diagnosis method, noise can be effectively suppressed, the bearing fault feature information can be effectively extracted, and the bearing state and the fault type can be accurately identified.

The invention relates to an internal solitary wave identification and amplitude inversion method in mode data based on machine learning, and belongs to the technical field of ocean observation. In combination with a Mask-RCNN and a variational auto-encoder, firstly, sea surface height data is preprocessed and imaged, texture features are extracted through a local binary pattern algorithm, a labeled data set is constructed, model training is performed, and internal solitary wave stripes and positions thereof are identified from the labeled data set; through extraction and processing of temperature field data, the temperature field at the water depth is inversed by using a variational auto-encoder; and extracting the amplitude of the internal solitary wave in the temperature field based on the internal wave identification result. According to the method, large-scale and refined ocean mode data can be efficiently and accurately processed, and the efficiency and accuracy of data processing are remarkably improved. The method has a wide application prospect, and especially in marine disaster early warning, marine scientific research and internal wave mechanics research, the automatic identification and amplitude extraction capability of internal waves can be greatly improved.

The invention relates to the technical field of crop disease and insect pest image recognition, and particularly discloses a crop disease and insect pest image recognition method based on a large model, and the method comprises the steps: obtaining a multi-angle leaf image through high-resolution imaging equipment under a controllable illumination condition, and obtaining a target image in a unified format; extracting scab texture complexity features in combination with a local binary pattern and a gray-level co-occurrence matrix algorithm, and performing multi-channel statistical analysis on RGB and HSV color spaces to generate color heterogeneity feature vectors; further fusing the two types of features into a composite disease feature vector, inputting the composite disease feature vector into a probability model constructed based on a support vector machine and a Monte Carlo Dropout mechanism, and outputting probability distribution and confidence score of disease and pest categories; and dynamically adjusting a model training strategy according to a confidence level, triggering a feedback mechanism for a low-confidence sample, generating a synthetic image by using a conditional generative adversarial network, and optimizing model parameters in combination with incremental learning to realize stable identification modeling of rare or complex disease types.

The invention discloses a method and a system for automatically identifying primordial follicles in an ovary microscopic section image, which are used for realizing automatic identification and counting of primordial follicles in the microscopic image by carrying out image binaryzation, expansion corrosion processing, edge detection, circular feature extraction and texture feature analysis on the ovary microscopic section image dyed under a microscope. According to the method, a local binary pattern (LBP) is adopted to extract texture features, and primordial follicles in a microscopic slice image are effectively distinguished by combining edge detection and a hollow circular screening algorithm; the method has the characteristics of high processing speed, high precision and strong adaptability, and can be widely applied to ovarian slice analysis, assisted reproductive medicine and research of ovarian function related diseases.

The invention discloses a discharge channel extraction method and system based on edge detection and texture feature fusion, and belongs to the technical field of power equipment fault diagnosis and digital image processing, and the method comprises the steps: obtaining an ultraviolet weak light image of power equipment, and carrying out the preprocessing; multi-scale edge detection is executed based on the preprocessed image, edge information is fused through a multi-scale voting mechanism, and a candidate edge graph is generated by combining direction consistency connection fracture edges; extracting texture features of the pre-processed image, wherein the texture features comprise a rotation invariant local binary pattern uniformity feature and a multi-direction gray level co-occurrence matrix feature; according to a dynamic distribution weight coefficient of a discharge type, carrying out weighted fusion on the edge intensity of the candidate edge graph and the texture features, and generating a discharge channel confidence score graph; and carrying out binarization segmentation and contour optimization processing on the confidence score graph of the discharge channel, and extracting morphology quantization parameters of the discharge channel.

The invention discloses a carpet printing quality detection and automatic correction method based on machine vision. The carpet printing quality detection and automatic correction method comprises the following steps: S1, collecting and preprocessing a carpet printing image; s2, inputting the preprocessed carpet printing image into a PSPNet network to obtain a regional feature map; s3, performing response weight increase on the low-frequency texture loss region and the high-frequency edge dislocation region by adopting a region response enhancement strategy to generate a defect thermodynamic diagram; s4, dividing a thermodynamic diagram and extracting weighted LBP (Local Binary Pattern) features; s5, carrying out clustering division on the weighted LBP features, and judging whether the candidate region is a printing defect region or not by using a metric function; s6, constructing a regional difference mapping graph, and calculating correction parameters; s7, a control signal is sent, and jet printing parameters are adjusted to execute correction; and S8, collecting the corrected image, repeating the steps S2 to S7, and judging that the quality reaches the standard. Intelligent closed-loop control from detection to correction is achieved, the printing precision and efficiency are improved, and the method is suitable for intelligent production of carpet patterns.

The invention discloses an ultra-high-definition video stream adaptive coding method based on deep learning visual saliency, and the method comprises the steps: carrying out the five-scale Gaussian filtering processing and image pyramid construction of a video frame, and combining Sobel gradient, Laplacian edge and local binary pattern feature extraction to generate a multi-scale feature map; a pre-training saliency detection network is adopted, and a smooth saliency thermodynamic diagram is generated through processing of a feature adaptation layer, a residual encoder, a self-attention mechanism and a transposed convolution decoder; dividing the video frame into a high region, a middle region and a low region according to the saliency thermodynamic diagram, and establishing a regionalization coding parameter table; performing differentiated prediction modes, motion estimation and quantization strategies on different salient regions; and organizing coded data according to an H.265 / HEVC standard, and embedding the saliency thermodynamic diagram into supplementary enhancement information for transmission. According to the method, the important region concerned by the user can be intelligently identified, a differentiated coding strategy based on content semantics is realized, and the coding efficiency is remarkably improved.

Disclosed here is a method of matching color and appearance of coatings containing at least one effect pigment. The method involves: obtaining, for a coating sample to be matched, sample color image data having pixel-specific color model properties; creating local binary patterns based on the pixel-specific color model properties; generating a binary pattern histogram for the local binary patterns, wherein the binary pattern histogram arranges the local binary patterns into N distinct bins to obtain N bin counts, and wherein the binary pattern histogram represents an appearance feature descriptor of the coating sample to be matched; and utilizing the binary pattern histogram for comparison against a previously generated binary pattern histogram that serves as an appearance feature descriptor of a candidate coating specimen.

The invention discloses a wood block surface flaw detection method based on image features, and belongs to the technical field of image data processing, and the method comprises the steps: obtaining an image of the surface of a target object, and carrying out the denoising and illumination normalization processing of the image, and obtaining a first image; performing wavelet transform decomposition to obtain edge distribution description data and shape description data corresponding to the first image, and generating geometric characteristic description; analyzing texture distribution of the first image in combination with a local binary pattern algorithm according to the geometric feature description, and generating a texture feature set; fusing the geometric feature description and the texture feature set to construct a multi-dimensional feature vector, and performing dimensionality reduction on the multi-dimensional feature vector by adopting principal component analysis to obtain a dimensionality-reduced feature vector; and inputting the dimension reduction feature vector into a support vector machine classifier, and outputting a defect classification result. The wood block surface flaw detection method based on the image features solves the problem that when an existing detection system faces complex flaw types, high-precision classification is difficult to achieve.

The invention relates to the technical field of machine vision detection, in particular to an automobile part surface defect online detection method and system. The method comprises the following steps: acquiring a grayscale image, calculating the size of a structural element for each pixel based on a local Gaussian Laplacian operator response variance, filtering to obtain a substrate image according to the size of the structural element, and differentiating to obtain a texture image. Determining a Gabor scale and a gray-level co-occurrence matrix statistical direction by using the size, and extracting a cooperative direction gray-level co-occurrence matrix feature; and a weight is set based on the size and is subjected to weighted fusion with a multi-scale rotation invariant local binary pattern feature to generate a texture saliency map, and texture defects are judged. On the substrate image, taking the gray value as the height, and determining a neighborhood calculation curvature feature based on the size to detect the substrate defect. According to the scheme, the image scale can be adaptively analyzed, the background texture is effectively inhibited, and therefore different types of tiny defects such as scratches and pits can be reliably detected.

The invention relates to the technical field of oil and gas field databases, and discloses a data cross-modal association method and device in an oil and gas field database, equipment and a medium, and the method comprises the steps: receiving an original seismic waveform time sequence inputted in advance, and generating a time-frequency spectrogram through Fourier transform; receiving a pre-input rock core scanning image set, extracting local binary pattern texture features from the rock core scanning image set, and obtaining a spatially continuous three-dimensional rock core digital model based on the local binary pattern texture features; injecting the time-frequency spectrogram and the three-dimensional rock core digital model into a pre-generated hierarchical semantic framework to obtain a cross-modal unified semantic vector; constructing a comparative learning framework of the spectrogram and the logging text, and generating a time-space aligned semantic feature matrix based on the comparative learning framework; and based on the cross-modal unified semantic vector and the semantic feature matrix, obtaining a dynamic association map of the logging and the seismic waveform.

The invention discloses a character extraction method for a power transmission and transformation project document. The method comprises the following steps: acquiring data information of a target power transmission and transformation project document and preprocessing the data information to obtain a binary image of the target power transmission and transformation project document; noise detection, noise classification and noise reduction processing are carried out; carrying out character region positioning on the target power transmission and transformation project document; segmenting and extracting characters; character recognition is carried out based on a local binary pattern; and carrying out association and replacement of characters and completing character extraction of the target power transmission and transformation project document. The invention also discloses a system for realizing the character extraction method for the power transmission and transformation project document. According to the method, the image of the target power transmission and transformation project document is processed and extracted, and particularly the characters of the table part in the document are positioned, segmented, identified and replaced, so that the character extraction of the power transmission and transformation project document is realized, the reliability is higher, and the accuracy is better.

The invention provides a mountainous area night animal tracking method and system based on multispectral and thermal imaging data fusion, and relates to the technical field of night animal tracking, and the method comprises the steps: synchronously obtaining a multispectral image sequence and a thermal imaging image sequence of a mountainous area night; performing multi-scale Retinex enhancement processing on the near-infrared image and the thermal image after registration to obtain an enhanced candidate target image; extracting a multi-source texture signature fusing a local binary pattern feature vector and a thermal gradient direction histogram for each candidate target, and constructing a multi-dimensional feature set in combination with geometric features obtained by morphological skeletonization; target association between adjacent frames is realized by adopting a nearest neighbor greedy matching algorithm, and a preliminary track segment is formed; and finally, removing abnormal trajectories, solving trajectory crossing and shielding conflicts through a sparse dynamic time warping algorithm, and generating an accurate and complete animal movement trajectory. According to the invention, conflicts caused by track crossing and shielding can be solved, and a continuous and accurate single animal movement track is output.

The invention discloses a sandstone slice image component identification method based on a GL-SLIC algorithm and semi-supervised learning. The method comprises the following steps: firstly, acquiring a sandstone slice image amplified by 200 times under single polarization, converting an RGB image into a CIE-Lab color space, and extracting a GLBP texture feature vector; then a GL-SLIC superpixel segmentation algorithm is constructed, and adaptive segmentation based on texture and color features is realized in combination with a Gabor filter and a local binary pattern; then implementing a region merging algorithm to generate complete mineral particles and pore fragments; finally, a classifier based on VGG16 and a discriminator based on ResNet18 are constructed, a semi-supervised self-training framework is adopted, a model is initialized by using about 6% of manual labeling samples, a high-confidence-coefficient pseudo-label extension training data set is generated through iteration, and automatic recognition of sandstone components such as quartz, pores, kaolinite, rock debris and a matrix is achieved. According to the method, the recognition accuracy of 96.3% on a test set is achieved, compared with a traditional method, the segmentation precision and the recognition accuracy are remarkably improved, the data labeling cost is greatly reduced, and an efficient technical scheme is provided for automatic analysis of geological images.

The invention provides an image classification method and device based on super-dimensional calculation, a medium and a product. The method comprises the following steps: extracting target texture features from an infrared image by using a local binary pattern; performing feature dimension reduction on the target texture features through a principal component analysis method to obtain one-dimensional structure feature information of the target; performing super-dimensional coding on the one-dimensional structure feature information of the target by using an efficient multivariate coding method to form a super-dimensional vector; constructing a super-dimensional vector data set by using the obtained super-dimensional vectors; constructing a target classification and recognition model based on a binary full-connection neural network; training a target classification recognition model by using the super-dimensional vector data set; and performing target recognition by using the trained target classification recognition model to obtain a target category. Target recognition is completed through combination of feature extraction, efficient super-dimensional coding and the binary full-connection neural network, the image classification recognition accuracy can be improved, and the method has the advantages of being low in super-dimensional coding complexity, high in model generalization ability and the like.

The invention discloses an AI linkage data system based on Internet of Things multi-source data fusion, and relates to the technical field of image recognition, and the system comprises a collection module which is used for collecting a real-time image of a fire, and selecting a suspected region with the brightness higher than a brightness threshold value in the real-time image; the preprocessing module is used for performing Gaussian filtering preprocessing on the suspected area; the method comprises the following steps: denoising a suspected region by dynamically adjusting Gaussian filtering parameters, calculating a brightness change rate of pixels on the inner and outer sides of an edge to obtain a gradient band width, dividing suspected region sub-blocks, extracting sub-block texture feature vectors by adopting a local binary pattern algorithm, and calculating a region overall texture entropy value by combining a region texture similarity matrix to obtain a region texture similarity matrix; the difference between the strong reflection light spots and real flames can be accurately recognized, the AI chip is prevented from misjudging the strong reflection light spots as newly added fire behaviors, then the linkage control module is prevented from mistakenly starting a spraying system in a non-fire behavior area, and damage of a large number of materials due to water stains caused by mistaken spraying is reduced.

The invention discloses a bridge concrete apparent damage identification method and system based on texture analysis, and belongs to the crossing field of bridge engineering detection and computer vision technology, and the method comprises the steps: employing an unmanned plane to collect a bridge member surface image, and generating an orthoimage; extracting contrast, correlation and entropy features based on the gray-level co-occurrence matrix; extracting histogram features based on a local binary pattern; detecting a corrosion dialysis area based on an HSV color space; fusing the multi-dimensional features, inputting the fused multi-dimensional features into a multi-label classifier, and outputting pixel-level segmentation masks of various damages such as cracks, spalling, exposed reinforcement corrosion and dialysis at the same time; according to the method, the complementary texture features are fused to realize simultaneous identification of multiple types of damages, a pixel-level segmentation result is output, and an evaluation report conforming to engineering specifications is automatically generated.

The invention relates to the technical field of highway engineering evaluation, in particular to a highway engineering quality monitoring and evaluating system based on an intelligent unmanned aerial vehicle. Defect classification is carried out through a convolutional neural network model, defect types are judged, crack detection, leakage detection and peeling detection are carried out on classified tunnel images, multi-direction gradients are introduced to calculate gradient intensities in different directions, fusion is carried out, and a fusion result is obtained. Jointly detecting a crack edge image in the tunnel image according to a fusion result and a local binary pattern, and calculating a crack width and a crack length based on the crack edge image; constructing a tunnel scoring model based on detection results of crack detection, leakage detection and spalling detection to calculate a tunnel quality score and a tunnel quality interval; and based on the current tunnel quality score and the collection time point, calculating a tunnel quality score and future risk time corresponding to the tunnel quality score in a subsequent level, and planning the latest collection time point of the unmanned aerial vehicle for collecting the tunnel image according to the future risk time.

The invention provides a video coding method, which comprises the following steps of: inputting a coding tree unit to be coded, and extracting texture features of a coding block, the texture features comprising local binary pattern similarity of all pixel points in the coding block and direction dispersion of each pixel point; inputting the texture features into the trained multi-type division prediction architecture, and sequentially judging whether to execute quadtree division, horizontal or vertical division, horizontal binary tree or ternary tree division and vertical binary tree or ternary tree division or not through a binary classification problem method; and calculating the rate distortion cost under each division mode to obtain an optimal division mode and a suboptimal division mode of the coding block, and outputting an optimal mode or a combination of the optimal mode and the suboptimal mode. According to the invention, the accuracy of the prediction mode in coding is improved.

The present invention relates to a wire rope surface defect recognition method based on feature fusion, which is in the field of wire rope surface defect recognition technology. The method comprises the following steps: graying a wire rope defect image, dividing the image into blocks, improving a traditional LBP algorithm using a central multi-scale local binary pattern based on the image blocks, extracting texture feature information of the divided image, performing PCA dimensionality reduction based on the obtained image texture features, and finally extracting global image texture features using GLCM and performing feature fusion with the reduced image texture features; and performing wire rope surface defect recognition and classification using an SVM classifier. The method solves the problem that traditional local binary patterns (LBP) are easily affected by central pixels and noise and cannot accurately identify wire rope surface defects. The method achieves an overall recognition rate of 97.5% for wire rope surface defects, which is at least 5% higher than other algorithms. The method can effectively identify various defects on the wire rope surface.

The invention discloses a target tracking method based on feature correlation and a dynamic model, which comprises the following steps: firstly, converting an image acquired by a tracking system into a Lab color space, calculating a color moment to describe target color distribution, extracting texture features by using a local binary pattern (LBP), and performing initial modeling by using a Gaussian mixture model in combination with an expectation maximization (EM) algorithm; target initial features are accurately obtained; then, predicting the motion state of the target based on a uniform motion model by adopting a self-adaptive particle filter algorithm, and improving the prediction accuracy; thirdly, candidate areas are divided in the prediction area, feature correlation matching is carried out by calculating the mahalanobis distance, and a target is accurately positioned; then, calculating speed and acceleration according to target position change, and dynamically updating model parameters by combining feature matching similarity to adapt to target dynamic change; and predicting the position by using particle filtering, and re-initializing the model to recover tracking after shielding is removed.

The present application provides an image processing method, device and equipment, which can be applied to the technical field of image processing. The image processing method comprises: pre-processing an input image to obtain input features; inputting the input features into a visual encoder and a multi-layer perception machine in a visual center decoupler respectively to obtain enhanced special features and salient object detection special features; the visual encoder aggregates local region features based on the input features to obtain the enhanced special features, and the multi-layer perception machine captures edge information based on the input features to obtain the salient object detection special features; inputting the enhanced special features into an enhancement network to obtain enhanced output features; the enhancement network takes illumination weights of different color channels and local binary pattern features of the input image as illumination constraints, and enhances the enhanced special features to obtain the enhanced output features; and inputting the salient object detection special features and the enhanced output features into a salient object detection network to detect a salient object.

The invention relates to a material processing method and system based on sensitive content detection and a medium, and belongs to the technical field of multimedia content security. The material processing method comprises the following steps: receiving an input original material, and analyzing the input original material into an image stream, a video stream and a text stream; executing HSV color space conversion, and extracting texture features; performing semantic analysis on the text stream to generate a word segmentation sequence and a named entity recognition result; random frequency noise is injected based on the saturation channel, and local binary pattern features are extracted from the saturation channel after noise injection; generating a semantic vector according to the word segmentation sequence and a corresponding named entity recognition result, and calculating a risk entropy value; and establishing a space-time mapping relationship between the image stream / video stream and the text stream, constructing a cross-modal incidence matrix, executing a grading decision according to an output result, executing a material processing operation, and feeding back a processing result to the cross-modal incidence matrix for weight updating. The sensitive content identification accuracy and timeliness can be improved.