12results about How to "Accurate border" patented technology

The invention belongs to the technical field of remote sensing detection, and particularly discloses a photovoltaic station fine identification method and device based on a classifier and deep learning, and the method comprises the steps: carrying out the preprocessing of a multi-source remote sensing image of a to-be-identified photovoltaic station in a research region range; inputting the preprocessed multi-source remote sensing image into a trained machine learning classifier to carry out preliminary coarse resolution recognition, and obtaining a high-resolution remote sensing image containing a coarse recognition contour of a potential photovoltaic region; and inputting the high-resolution remote sensing image containing the coarse identification contour of the potential photovoltaic region into a trained deep learning model for fine identification, and outputting a segmentation mask with panel type information, the segmentation mask being used for distinguishing a photovoltaic panel and a panel gap. According to the method, the potential area of the photovoltaic station can be quickly positioned in the specified area range, and fine identification is carried out on the basis, so that the accuracy and efficiency of identifying the photovoltaic station in a complex and diversified area are improved.

The invention relates to a state monitoring and fault diagnosis method based on a digital twinning sensing framework, which comprises an offline stage and an online stage, and the offline stage comprises the following steps: dividing a numerical calculation discrete model according to a physical structure of industrial equipment to be detected; based on the numerical calculation discrete model, determining a partial differential equation corresponding to the physical field, and collecting and obtaining a data set of the physical field; data analysis is carried out on the data set, real-time calculation of a high-dimensional physical field is realized, a reliability evaluation system, a physical filtering system and an abnormal working condition identification system are constructed respectively, and a digital twinning perception framework is formed; the on-line stage comprises the following steps of: acquiring actually measured data of industrial equipment to be measured, inputting the actually measured data into a digital twinning perception framework, and performing reliability evaluation, physical filtering and abnormal working condition identification. Compared with the prior art, the method is high in calculation efficiency, has interpretability, and can efficiently and accurately carry out state monitoring and fault diagnosis.

The invention discloses an improved U 2 The significant Mongolian pattern detection algorithm based on ‑Net includes: obtaining an image to be detected; inputting the image to be detected into a significant Mongolian pattern detection model, and outputting a significant Mongolian pattern detection result; wherein the significant Mongolian pattern detection model is based on an improved U 2 The deep learning model of the ‑Net network includes an encoder, a decoder, a feature supplement extraction module, a lightweight combined attention module and a feature map fusion module; the feature supplement extraction module adopts a small-size convolution kernel series connection method to increase the receptive field, reduce the computational cost, and make the predicted saliency map more complete; the lightweight combined attention module makes the boundary positioning of the saliency map more accurate without significantly increasing the computational cost, while suppressing redundant background noise, solving the technical problems of incomplete prediction of saliency Mongolian patterns, inaccurate edges and low detection accuracy.

The present application relates to a kind of structure inference, automatic analysis and standardized output of multi-variant BOM table, belong to electronic manufacturing data processing, table analysis and structured data cleaning field.The present application constructs a set of analysis architecture for multi-variant BOM table by the modular processing link of "title row detection-column head normalization-multi-column identification-variant reconstruction-segment cleaning-intelligent inheritance-standardized output", robustly locates complex table head by strong and weak evidence weighting and verification under-probing mechanism, realizes the split and identification of multiple quantity columns using stop mark and semi-finished product number analysis, and introduces cross-row backfilling and position number inheritance rules under paragraph constraint, effectively solves the analysis problem caused by template difference, multi-column variant and information loss, realizes the automation, precision and standardization of BOM data from messy input to engineering level standardized output, significantly reduces manual intervention and improves data quality.

The invention discloses an automobile-oriented dynamic electronic fence generation method, which comprises the following steps of: acquiring vehicle information, denoising, mining a risk area and a resident place, and generating an initial risk area model with boundary coordinates and risk levels; then, administrative division and road vector data are fused, a three-dimensional hierarchical fence structure is constructed, and fence boundaries are optimized through a Douglas-Peucker algorithm and a cubic B-spline function; according to the real-time driving state of the vehicle, the fence radius is dynamically adjusted through a linear expansion algorithm, the alarm sensitivity is regulated and controlled in a graded mode, and dynamic fence configuration is generated; and finally, vehicle positioning is monitored based on a state machine model, fence switching and boundary jitter are processed by combining priority arbitration and a hysteresis comparison mechanism, redundant alarms are filtered through a time sliding window and behavior intention analysis, and an early warning report with a context is generated. According to the invention, dynamic and multi-level management and control of the electronic fence are realized, the accuracy and suitability of vehicle risk control monitoring are improved, and the false alarm rate is reduced.

The application discloses a remote sensing semantic change detection method based on mixed Mamba-Transformer and cross attention fusion, inputs a double-time-phase remote sensing image into a weight-shared mixed Mamba-Transformer encoder to extract multi-level double-time-phase features; a complementary space-time interaction fusion module performs space-time projection, parallel Mamba scanning and cross attention calculation on the features to generate change perception fusion features; a change detection decoder combines the fusion features and the multi-level features to output a binary change mask, and two parallel semantic segmentation decoders generate double-time-phase land cover semantic probability maps respectively; and the change mask is used to constrain the semantic probability maps in space to obtain a semantic change detection result. The application considers global modeling efficiency and local detail description ability through a mixed architecture, effectively suppresses environmental pseudo changes through parallel Mamba and cross attention fusion, and guarantees the consistency of change areas and semantic categories from the mechanism through cross-task collaborative decoding, so that the robustness and precision of remote sensing semantic change detection in a complex environment are significantly improved.

The invention belongs to the field of image region segmentation, and particularly relates to an AI-based colorectal tumor image region segmentation method and system, and the method comprises the following steps: S1, obtaining original 3D colorectal tumor image data; according to the method, the high-quality 3D colorectal tumor image data is processed through the surface parameterization technology, the obtained 2D feature map is more suitable for processing the hybrid FMU-attention network model, and the processing efficiency and the processing speed of the hybrid FMU-attention network model are greatly improved; in addition, an AI framework is formed by jointly fusing a hybrid FMU-attention network model, a Vision Transformers module and a DeepLabV3 + model, the AI framework enables the boundary to be more accurate and the multi-scale performance to be more excellent, and it is ensured that the tumor can be effectively recognized and segmented regardless of the size.

The embodiment of the invention discloses an ultrasonic elastography method and ultrasonic imaging equipment. The ultrasonic imaging equipment determines a stress result according to the shear wave elasticity reference result and the strain elasticity reference result, and determines a first quantitative elasticity result of the target area according to the stress result and the strain elasticity result corresponding to the target area; and determining a second quantitative elasticity result of the first region of interest according to the shear wave elasticity result of the first region of interest and the first quantitative elasticity result of the target region, and displaying the second quantitative elasticity result. Since the shear wave elasticity reference result, the strain elasticity reference result and the strain elasticity result of the target area are accurate, the first quantitative elasticity result and the second quantitative elasticity result calculated according to the above results are more accurate than the results obtained by directly performing shear wave elasticity measurement on the target area; and the hardness condition of the tissue can be described more accurately.

The application discloses a structure terrain semantic segmentation image optimization method based on gradient data. The method comprises the following steps: converting depth data (Depth) into gradient data with image pixel gray value change based on a Sobel operator, and normalizing the gradient data to show a normal map with red, green and blue (RGB) data; fusing the normal map and the red, green and blue (RGB) data according to a pixel point matrix Max method, re-encoding the input image, and outputting image information with color representation of normal direction information and basic RGB image fusion; and based on the Canny edge detection data and the gradient data of the normal map, optimizing the edges of the semantic segmentation result of the MobileNetV3 lightweight network. Compared with the prior art, the method optimizes the edges of the semantic image, and simultaneously considers speed and accuracy. The semantic segmentation edge optimization method is simple and efficient, and is suitable for semantic segmentation of most structure terrains.

The invention relates to the technical field of image analysis, in particular to a leakage point detection method in a waterproof leakage treatment process, and the method comprises the steps: obtaining a plurality of blocks in a gray-scale map of a thermal infrared image of a leakage region to be detected, and then screening out edge blocks, inner blocks and background blocks in the plurality of blocks; performing region iteration adjustment on the edge blocks, the inner blocks and the background blocks to obtain adjusted edge blocks, inner blocks and background blocks, and obtaining the edge blocks according to the significance values of the adjusted edge blocks and the distribution conditions of the adjusted edge blocks of the adjusted inner blocks in all directions. Determining an adjustment significance value of the adjusted inner sub-block; and determining a leakage area based on the adjusted saliency values of the edge blocks and the background blocks and the adjusted adjusted saliency value of the inner blocks. According to the invention, the accuracy of leakage detection is effectively improved.

The application discloses a kind of dynamic electronic fence generation methods for car, this method first gathers vehicle information after denoising, excavates risk area and permanent site, generates initial risk area model with boundary coordinates and risk level;Again, administrative division, road vector data are fused, and three-dimensional hierarchical fence structure is constructed, and fence boundary is optimized by Douglas-Pork algorithm and cubic B-spline function;Then according to the real-time driving state of vehicle, the fence radius is dynamically adjusted by linear inflation algorithm, and the alarm sensitivity is controlled by classification, to generate dynamic fence configuration;Finally, based on state machine model monitoring vehicle positioning, combined with priority arbitration, hysteresis comparison mechanism handles fence switching and boundary jitter, filters redundant alarm through time sliding window and behavior intention analysis, generates warning report with context.The application realizes the dynamicization of electronic fence, multi-level management and control, improves the accuracy and adaptability of vehicle risk control monitoring, and reduces the false alarm rate.

The embodiment of the invention provides a three-dimensional scene text positioning method and device, equipment and a storage medium, and relates to the technical field of image processing. The method comprises the following steps: acquiring a depth map corresponding to each view angle by using a three-dimensional Gaussian model, and acquiring a semantic coding vector of at least one object pixel block corresponding to the view angle; encoding the obtained text description to obtain a text embedding vector, for each view angle, obtaining a similarity matrix based on the text embedding vector and a semantic encoding vector, selecting an object pixel block corresponding to the maximum similarity in the similarity matrix as a target pixel block, and obtaining a three-dimensional center position corresponding to the target pixel block based on the depth map; and performing multi-view clustering voting on the three-dimensional center positions corresponding to all the views to obtain a multi-view center position, and obtaining a view cone region of the target object based on the multi-view center position. The method improves the accuracy of a positioning result, does not need to train scenes one by one, expands the application range of three-dimensional scene text positioning, and is especially suitable for online scenes.