27 results about "Computer image" patented technology

This invention discloses a microscopic denoising method based on multi-expert judgment, relating to the field of computer image enhancement technology. The method includes: acquiring original microscopic image data; preprocessing the image data; calculating initial expert weights through an expert weight threshold discrimination network; inputting the preprocessed data into multiple expert denoising networks for denoising; calculating result weights based on the denoising results and performing weighted fusion; and generating a denoised microscopic image. By introducing a multi-expert structure and threshold discrimination mechanism, this invention can adaptively denoise under various observation objects and imaging conditions, significantly improving the generalization ability of the denoising method. This invention, through a multi-expert mechanism, overcomes the shortcomings of current denoising methods, such as weak generalization and applicability limited to specific observation sample types or imaging conditions.

The application discloses a CBCT metal artifact suppression method based on an unsupervised preoperative prior repair network and belongs to the technical field of computer image processing. The method comprises the following steps: inputting a preoperative prior CBCT image and a CBCT image to be repaired into a generator after registration, respectively extracting anatomical structure features and non-metal region detail features by using two encoders of a double-flow feature extraction network, fusing the features, and then performing repair reconstruction on the CBCT image by using a decoder; inputting the result into a discriminator to perform authenticity discrimination and metal region discrimination and generate a fused prediction image; taking the prediction image information as a supervision signal to optimize the adversarial loss between the generator and the discriminator and the image similarity loss between the repaired image and the image to be repaired, and performing joint adversarial training on the discriminator and the generator; and processing an intraoperative CBCT image to be repaired by using the trained generator to realize more efficient metal artifact suppression.

The application provides a method for distinguishing the color of sub-group tobacco leaves, applied to the technical field of computer image processing, and comprises the following steps: obtaining a color classification sample image set of sub-group tobacco leaves, performing binary segmentation on the color classification sample of the tobacco leaves, and obtaining a binary image tobacco leaf area; performing coordinate return on the binary image tobacco leaf area, and generating a color image tobacco leaf area; extracting a pixel point Lab value set of the color image tobacco leaf area, combining a pure color pixel point Lab value, and calculating a pixel point contrast set; performing interval division on the pixel point contrast set, and generating a contrast interval division result; traversing the division result, and calculating an interval pixel proportion set; according to the interval pixel proportion set, drawing a proportion threshold point line graph, and then performing color distinction on a to-be-tested tobacco leaf image according to a voting mechanism. The method solves the technical problems that there is no standardized method for distinguishing the color of sub-group tobacco leaves in the prior art, manual classification has low efficiency and low classification accuracy, and is highly subjective.

A camera tracking system receives patient reference tracking information indicating pose of a patient reference array tracked by a patient tracking camera relative to a patient reference frame. A local XR headset view pose transform is determined between a local XR headset reference frame and the patient reference frame. Remote reference tracking information is received indicating pose of a remote reference array tracked by a remote reference tracking camera. A remote XR headset view pose transform is determined between a remote XR headset reference frame of a remote XR headset and the remote reference array. A 3D computer image is transformed from a local pose determined using the local XR headset view pose transform to a remote pose determined using the remote XR headset view pose transform. The transformed 3D computer image is provided to the remote XR headset for display with the remote pose relative to the remote XR headset reference frame.

This invention relates to the fields of artificial intelligence and computer image recognition technology, and discloses an intelligent method and system for detecting the quality of interior decoration. The method includes: applying a composite acoustic signal through a multi-band acoustic excitation device, and simultaneously acquiring multi-channel acoustic response, infrared thermal imaging sequences, and electromagnetic induction data; extracting acoustic impedance abrupt changes, thermal conduction anomalies, and electromagnetic continuity features respectively; generating multimodal embedding vectors using three convolutional neural network encoders, and generating a unified representation through a cross-modal attention fusion module; finally decoding and outputting defect type, confidence level, and three-dimensional positioning information, and superimposing it onto a building information model to generate a visual report. This system integrates a multi-physics sensing unit and an edge computing module to achieve non-invasive, fully automatic, and high-precision interior decoration quality detection. This invention improves defect detection rate and positioning accuracy through multimodal fusion and deep learning, increasing detection efficiency compared to manual methods.

The present application relates to the field of computer image processing, in particular to a feature enhancement and self-adaptive regression tea leaf picking point positioning method and system. The basic principle of the method is: firstly, the collected tea bud image is preprocessed for clarity; then, a two-stage model architecture of detection first and then positioning is adopted, an improved YOLOv5 network is used in the detection stage to robustly detect multi-scale buds and suppress background interference; then, the target suitable for picking is screened out and cut into a single bud image; in the positioning stage, an improved YOLOv11-Pose network integrated with an adaptive convolution kernel module is used to accurately regress the picking point coordinates; finally, the coordinates are mapped back to the original image and output. The core technical effect of the present application is: through the synergistic optimization of the two-stage process and the targeted improvement of the model components, the problem of inaccurate picking point positioning and poor robustness caused by image degradation, multi-scale targets, complex background and variable bud morphology in the natural environment is effectively solved, providing a high-precision solution for tea leaf automatic picking.

The utility model discloses a kind of determination device of carbonyl iron in coal-based methyl acetate, including ultraviolet light source controller, the ultraviolet light source controller is connected to multiple ultraviolet lamp light sources by signal line, ultraviolet light source controller can simultaneously adjust control wattage, wavelength and luminous flux of multiple ultraviolet lamp light sources;Ultraviolet lamp light source is used to irradiate high-transmittance analysis dish, and special coating mirror surface passes reaction image in high-transmittance analysis dish to high-speed camera, and high-speed camera is connected with computer image analysis system.The utility model can realize multiple sample parallel detection, with the characteristics of small error, accurate control.

This invention discloses a method and system for constructing a disaster resilience classification training program for clinical nurses, belonging to the fields of disaster nursing management and computer image recognition technology. It involves acquiring facial micro-expression image sequences of nurses in standardized disaster simulation scenarios; extracting the activation frequency, activation duration, and left-right facial activation symmetry features of the zygomaticus major and orbicularis oculi muscles and inputting them into a preset resilience classifier; outputting subtype labels based on a fixed mapping relationship between the spatiotemporal features of muscle activity and resilience subtypes; retrieving the corresponding basic training framework based on the labels; comparing the activation duration features with a threshold; and generating reinforcement training labels on the framework using image overlay technology to form a personalized classification training program. This invention achieves objective and real-time subtype identification of clinical nurses' disaster resilience and can automatically generate differentiated training content targeting resource deficiencies in different subtypes, significantly improving the accuracy and automation level of disaster resilience training.

The application relates to the technical field of computer image segmentation, and provides a fuzzy boundary segmentation method based on model guidance enhancement, which comprises the following steps: replacing the skip connection layer of a standard UNet model with a boundary extraction module, a multi-task learning module and a cross-feature fusion module which are connected in series, and adopting the final segmentation result to constrain the boundary semantic information extracted by each layer of the encoder of the standard UNet model to obtain a fuzzy boundary segmentation small model; adopting a SAM-Med2D large model to predict an input image to obtain a segmentation image; adopting the fuzzy boundary small model to predict the input image to obtain a boundary image; adopting a threshold value self-adaptive mutual enhancement mechanism to output a prompt-enhanced final segmentation result; and the method improves the precision and robustness of image boundary segmentation.

The present application relates to the field of geotechnical engineering and computer image processing technology, and particularly relates to a residual soil structure weakening area identification method based on CT images. The method comprises the following steps: constructing a distance field map for the residual soil CT image; segmenting independent particle objects based on the distance field map, and establishing a particle instance index map; identifying a contact connection area based on the particle instance index map; extracting the effective neck width of the contact connection area according to the distance field map, and correlatively calculating the equivalent geometric diameters of the particle objects on both sides of the contact connection area to construct a contact feature data set; and calculating the area contact bottleneck coefficient of the contact connection area based on the contact feature data set. The present application quantifies the geometric bottleneck features of the particle connection and analyzes the local topological stability, realizes the quantitative identification and spatial visual positioning of the weak areas in the internal structure of the residual soil, and effectively identifies the particle connection structure which is geometrically connected but mechanically unstable.

The application provides an underwater vehicle terrain scanning process simulation method, system and medium, and the method comprises the following steps: constructing a multi-beam sonar model; constructing an underwater three-dimensional terrain, and determining a navigation scheme of the underwater vehicle; applying the multi-beam sonar model to the navigation scheme to perform imaging operation, and obtaining an underwater vehicle terrain scanning process model; adding multi-source interference to a terrain scanning result of the underwater vehicle terrain scanning process model to generate a comprehensive simulation image; and visualizing a scanning progress of the underwater vehicle terrain scanning process model, and synchronously outputting the scanning result in the scanning progress; the method can simulate an actual process, provide an experimental platform, reduce experimental cost, provide a guidance scheme and verification for actual detection work, improve work efficiency, and the comprehensive simulation image after adding the multi-source interference has a very realistic effect, can provide data support for an image processing algorithm, and can be widely applied to the technical fields of underwater sensing and computer image processing.

This invention belongs to the field of computer image processing and computational wood science, specifically relating to an intelligent batch identification and analysis system for cells in cross-sections of gymnosperm wood. The system includes sequentially connected modules for image input, preprocessing, cell instance segmentation, unique ID generation and management, batch calculation of geometric parameters, and result output and interactive filtering. The method involves: after preprocessing the input image, using instance segmentation based on traditional image processing algorithms to batch identify cell outlines; assigning a unique ID to each cell and establishing a mapping; calculating the cell lumen area, diameter, and wall thickness in batches based on the mapping relationship; and finally, performing visualization and interactive filtering. This invention, by integrating batch parallel processing and cell instance-ID mapping management, achieves minute-level fully automated identification and measurement of massive numbers of cells in wood cross-sections, improving efficiency by hundreds of times compared to manual serial operations, and supports fine-grained data management and statistics.

This invention relates to the interdisciplinary field of computer image detection and biomedical engineering, and discloses a method and system for respiratory motion monitoring based on Kinect. The method includes: simultaneously acquiring Kinect 3D point cloud sequences and bioimpedance time-series data; extracting first and second respiratory signals respectively; evaluating Kinect signal quality in real time, and generating continuous respiratory waveforms using a main signal output or dynamic weighted fusion strategy under non-interference, interference, and transitional states respectively; and calculating respiratory rate, relative rate of change of tidal volume, and rhythm variation coefficient. The system includes a Kinect sensing unit, a bioimpedance measurement unit, a signal processing and fusion unit, and a respiratory parameter calculation unit. This invention significantly improves the robustness and reliability of respiratory monitoring in complex scenarios through dual-modal redundant sensing and adaptive fusion mechanisms.

The application discloses a kind of isolated block curved surface axial angle measurement method, device, equipment and medium, applied to part angle measurement field, comprising: obtaining isolated block concave curved surface 3D depth image;Depth image is carried out gray processing, type conversion and specification conversion, and obtain pre-processing image;Region extraction is carried out to pre-processing image, and obtain the effective area containing isolated block curved surface;The boundary size of effective area is measured, and longitudinal median axis is determined according to boundary size;Directional line in effective area is screened, and target straight line is determined;The included angle of target straight line and longitudinal median axis is calculated to obtain isolated block curved surface axial angle.The application utilizes computer image processing technology, i.e.mechanical vision, to carry out automatic detection method, compared with artificial detection, can greatly reduce the detection labor intensity of worker, and detection precision is high, fast, and the qualified rate of product can be effectively guaranteed.

The application discloses a kind of weld image preprocessing methods for improving edge sampling precision, belong to computer image processing and artificial intelligence field.It includes: step one, weld image is carried out gray processing and contrast enhancement, as enhanced picture;Edge detection is carried out to enhanced picture, and edge feature map is obtained;Step two, the edge feature map obtained in step one is carried out gray mapping and weld material noise elimination, and edge feature binary graph is obtained;Step three, based on the geometric relation of weld edge position and sliding window sampling frame, the height of weld photo is adaptively adjusted, and the complete collection of edge information is realized when sliding window sampling is carried out.The method is suitable for computer intelligent edge recognition and macro parameter analysis of steel pipe double-sided submerged arc weld, and the complete collection of edge information can be realized when superpixel sampling is carried out, and the sampling precision of weld edge in sliding window sampling process is improved.

The present application relates to the field of computer image processing, and discloses a kind of image path generation method and device for engraving processing constraint, including input RGB image, pre-processes RGB image, and generates multimodal data;Gradient field is constructed according to multimodal data, structural guiding weight function is constructed based on structural saliency, each modality image in multimodal data is weighted and fused according to structural guiding weight function, and fusion image is obtained;The fusion image is input into the enhancement network for detail enhancement, and enhanced image is output;Path extraction and curvature calculation are carried out on the enhanced image, so that path set and curvature sequence are obtained;Path optimization is carried out according to the obtained path set and curvature sequence;The final PLT file is generated.The present application has the advantages of improving detail retention capability, improving path continuity, reducing file complexity, and directly meeting the requirements of engraving processing.

This invention relates to the field of computer image processing technology, and particularly to an optimized method for measuring the volume between intraocular implants and anterior segment structures. The invention utilizes anterior segment OCT grayscale images, preprocesses the OCT images using specific image processing methods, generates a high-resolution OCT grayscale image mask using a specific mask generation method, and combines mathematical methods to automatically identify, 3D reconstruct, and measure the volume between the posterior surface of the intraocular implant and the anterior surface of the lens in the image. This invention fills the gap in the fully automated measurement of the volume between intraocular implants and anterior segment structures based on anterior segment OCT grayscale image recognition, proposes a new image preprocessing workflow, optimizes the speed and accuracy of Mask RCNN for high-resolution anterior segment OCT grayscale image recognition, and simultaneously enables 3D reconstruction and automatic measurement of the spatial volume between the posterior surface of the intraocular implant and the anterior surface of the lens.

This application provides an automatic classification method and system for lunar impact craters, applicable to the field of computer image processing technology. This method addresses the technical problems of low accuracy and low robustness in lunar impact crater classification in related technologies. It includes: adaptively setting dual threshold parameters for an edge detection algorithm corresponding to the image signal-to-noise ratio (SNR) to obtain an adaptive edge detection algorithm for adaptive edge detection of a remote sensing single-channel grayscale image; performing gradient detection on the remote sensing single-channel grayscale image based on a preset first-order differential operator to obtain a fused edge image; adaptively generating fusion weights based on the image SNR; extracting features from the fused edge image and the remote sensing single-channel grayscale image respectively to obtain multi-dimensional feature data; and weighting and summing the normalized multi-dimensional features according to the dimensional weights adaptively generated based on the image SNR to obtain a comprehensive freshness score, thereby improving the robustness and accuracy of lunar impact crater classification.

This invention discloses a method, system, device, and medium for cross-domain target detection in traffic videos based on feature reconstruction, relating to the field of computer image data processing technology. The method includes the following steps: constructing a student model and a teacher model; acquiring a source domain image, inputting the source domain image into the student model, and calculating the source domain detection loss; acquiring a target domain image, inputting the target domain image into the teacher model to generate target domain pseudo-labels; inputting the target domain image and its corresponding pseudo-labels into the student model, and calculating the target domain pseudo-label detection loss; setting up a semantic feature reconstruction module to obtain the semantic feature reconstruction loss; setting up an instance feature reconstruction module to obtain the instance feature reconstruction loss; weighted summing of the source domain detection loss, target domain pseudo-label detection loss, semantic feature reconstruction loss, and instance feature reconstruction loss to form the overall training objective function; and after training, performing online inference on the road video stream to output the detection results.

The present application relates to the technical field of data processing. Disclosed are a method and apparatus for measuring the liquid level of a slag conveyor, a storage medium, and an electronic device. The method comprises: acquiring image data of a liquid level scale in a slag tank of a slag conveyor acquired by an industrial camera; by means of a bilateral filter, pre-processing the image data to obtain recognizable image data; inputting the recognizable image data into a preset deep learning model for image recognition, so as to obtain point information of the liquid level scale, the preset deep learning model being used to recognize the point information of the liquid level scale contained in the image; and, on the basis of the point information, determining a liquid level reading of the liquid level scale. Compared with the prior art, the present application can use computer image recognition technology to process real-time images, which greatly enhances work safety and reduces pollution and corrosion of measurement components by the environment while improving the measurement accuracy, thereby reducing labor and apparatus costs and also improving the accuracy of the measured liquid level.

This invention provides a method and system for rendering large-scene geographic vector data based on multi-camera collaboration and cascaded projection, belonging to the field of computer image vision technology. The method includes: loading and parsing the original geographic vector dataset, extracting primitive information of multiple primitives and creating corresponding graphic components; initializing multiple sub-cameras with different preset resolutions and configuring corresponding rendering parameters; dynamically determining and activating a target sub-camera from multiple sub-cameras based on the current viewpoint position information of the main camera; performing frustum clipping on the graphic components in the current rendering frame of the main camera to obtain the spatial bounding box information of the target component to be rendered; updating the camera parameters of the target sub-camera based on the spatial bounding box information of the target component and rendering it; and controlling the decal component to project the rendering result onto the 3D scene. This invention solves the problems of storage expansion, preprocessing time consumption, and unsmooth visual switching caused by LOD processing.

This invention relates to the field of tableware disinfection and packaging technology, specifically an intelligent tableware disinfection and packaging system. It aims to solve the problems of high workload and harsh working environments (such as high noise, high temperature, and high humidity) associated with manual tableware placement on packaging lines in tableware disinfection factories. In existing technologies, plates, bowls, cups, and other tableware, after being washed and disinfected by dishwashers, are mixed together, some with their rims facing up and others with their bottoms facing up. Traditional machinery struggles to effectively handle this complex situation, requiring manual placement and sealing, resulting in low efficiency and high labor intensity. This invention provides an intelligent machine that uses a computer image recognition program to classify and identify tableware on the production line, accurately determining the type and orientation of the tableware. It then combines a robotic arm with a flipping mechanism to complete the grabbing, flipping, and classification of the tableware. Addressing the issue of inconsistent orientation of tableware transported from the conveyor belt to the packaging line (e.g., tableware with its rim facing down needs to be flipped to face up), this invention designs a high-speed, low-cost flipping mechanism suitable for tableware of various shapes and sizes. By placing the tableware as required on the sealing line, automated assembly and sealing operations are completed. Compared with existing technologies, this invention significantly reduces manual labor intensity, improves production efficiency and hygiene standards, and is widely applicable to automated packaging processes in tableware sterilization factories.

This application discloses a character skin rendering method, apparatus, electronic device, and storage medium, relating to the field of computer image rendering technology. The method includes: acquiring a curvature map, a profile ID map, and multiple different diffuse profiles; performing curvature calculation processing on each diffuse profile based on the curvature map and the profile ID map to obtain representative curvature parameters for each diffuse profile; performing pre-integration calculation processing on each diffuse profile using the corresponding profile function and representative curvature parameters to generate multiple one-dimensional pre-integration results; constructing an integrated lookup table based on the multiple one-dimensional pre-integration results; and during the rendering process, performing character skin rendering processing on the character model based on the profile ID map and the integrated lookup table to obtain the character skin rendering result. This method can balance lower performance overhead with better skin rendering effects.

The application relates to the technical field of computer image processing, and discloses an expression animation display method and system, the method comprising the following steps: when expression animation is displayed, a control instruction sent by a central control host is acquired, the control instruction is analyzed, and analyzed control information is obtained; whether the expression animation is upgraded is determined according to the control information; when it is determined that the expression animation is not upgraded, whether the expression animation is switched is determined according to the control information; when it is determined that the expression animation is switched, an expression resource animation in a TTF font file format is acquired from a character set; font picture switching is performed according to the expression resource animation, and screen display is adapted. The expression animation is displayed in a vector font TTF display animation picture and a Bluetooth service analysis instruction control expression animation upgrade display mode, so that the smoothness of expression animation display and the dynamic adaptation capability can be ensured, and the real-time update requirement of expression animation resources in an MCU system can be met.

The application provides an infrared small target detection method of an asymmetric feature enhancement network, relates to the technical field of computer image segmentation, and comprises the following steps: pre-processing an input infrared image to represent a fixed-size feature tensor; defining a mixed convolution residual module, extracting image basis and directional enhancement features through double-branch convolution; adopting an asymmetric padding strategy for a spin-leaf type convolution to complete convolution and aggregation; defining a feature weighted fusion set, realizing double-branch feature fusion through a fusion weight alpha; combining an activation function, an attention mechanism and a residual connection processing to fuse features; continuously performing an encoder feature extraction operation on the features to screen small target discriminative features; obtaining intermediate layer features through a pooling convolution, and sequentially sampling and splicing multi-scale features through a decoder; generating multi-scale prediction maps and unifying resolutions, combining a gating mechanism to weight and fuse to output a target mask; and strengthening expression of infrared small target shallow features, accurately fusing multi-scale features and suppressing background interference.

This invention discloses a few-sample image classification method based on multi-granular semantic priors and semantically guided feature enhancement, belonging to the field of computer image classification. This method constructs a semantic prior for few-sample learning by introducing multi-granular semantic information at both the category and image levels. It then uses category-level semantic similarity to filter and fuse the basic category priors, thereby calibrating the prototype of the new category. Based on this, an image-level semantically guided feature enhancement mechanism is employed to uniformly enhance the visual features of support samples and query samples, improving the consistency and discriminative power of feature representations. This invention can effectively alleviate the prototype estimation bias problem under conditions of extremely small sample sizes, improving the stability and generalization performance of few-sample image classification, and has good application value.