83 results about "Imagery analysis" patented technology

The invention relates to the technical field of image analysis and mode recognition, in particular to a state monitoring and fault diagnosis method and system based on artificial intelligence. The invention discloses a state monitoring and fault diagnosis method based on artificial intelligence. The method comprises the following steps: S1, acquiring oil sample type data, performing background purification processing, and then executing a high-precision panoramic imaging process; the X-direction shooting number and the Y-direction shooting number are obtained based on the two-axis motion platform; s2, generating a grid coordinate sequence covering the deposition area according to the X-direction shooting number and the Y-direction shooting number, and forming a line-by-line scanning path; and synthesizing a full-focal-plane image based on the scanning path, and outputting a fused panoramic depth panorama according to the synthesized full-focal-plane image. According to the invention, through background purification, panoramic imaging, improved Mask-R-CNN mode recognition model segmentation and multivariate index fusion, precise recognition of wear particles and intelligent early warning and grading of wear states are realized.

The invention belongs to the technical field of hydraulic engineering monitoring, and discloses a hydraulic engineering inspection system and method, and the method comprises the steps: a positioning monitoring deployment module reasonably distributes the number and positions of positioning monitoring assemblies; the route planning module plans an optimal unmanned aerial vehicle flight route; planning an advancing route, a water entry position and a water surface leaving position of the underwater robot; the cruise monitoring module carries out all-directional inspection on a water conservancy area; the image analysis module identifies abnormal conditions; the infrared data analysis module recognizes abnormal conditions in time; the radar data analysis module identifies abnormal conditions; the comprehensive analysis module comprehensively evaluates the water conservancy condition; the risk prediction module predicts potential problems and risks. According to the invention, the positioning monitoring assembly and the cruise monitoring module are combined to realize omnibearing monitoring of a water conservancy area; omnidirectional three-dimensional inspection is realized, multi-modal data collaborative analysis is realized, abnormal conditions are identified, potential problems and risks are predicted, and possible faults or disaster hidden dangers of water conservancy facilities can be found in advance.

Systems and methods for performing a medical imaging analysis task using a foundation model are provided. One or more 3D (three-dimensional) medical images each comprising a plurality of 2D (two-dimensional) slices are received. A first set of features is extracted from the plurality of 2D slices of the one or more 3D medical images using a machine learning based encoding network. Each respective feature of the first set of features is resampled based on a spatial location of one or more pixels of the 2D slices from which the respective feature was extracted. The resampled first set of features is encoded into a second set of features. A medical imaging analysis task is performed based on the second set of features. Results of the medical imaging analysis task are output.

The invention provides a breast cancer-oriented traditional Chinese medicine syndrome type intelligent identification method and system, a terminal and a medium, and relates to the technical field of artificial intelligence and traditional Chinese medicine tongue diagnosis image analysis. Inputting the tongue picture image into a pre-constructed tongue picture element classification model to obtain a tongue picture element classification result; acquiring pulse condition data and clinical symptom data of the breast cancer patient; and according to the tongue picture element classification result, the pulse condition data, the clinical symptom data and a plurality of preset candidate syndromes, utilizing a preset matching function to calculate a matching score corresponding to each preset candidate syndrome, and taking the preset candidate syndrome with the highest matching score as a final syndrome. According to the invention, the analytic performance of intelligent recognition of the breast cancer-oriented traditional Chinese medicine syndrome type can be improved. Wherein the tongue picture element classification model is improved on the basis of a YOLOv8n-cls network architecture, the tongue picture classification accuracy can be improved, and meanwhile, the model deployment cost is effectively reduced.

Systems and methods for performing a medical imaging analysis task conditioned on multi-domain medical images with missing modalities are provided. 1) one or more medical images each in a different domain and 2) a domain code defining a presence of the different domains in a set of predefined domains are received. One or more weights are determined based on the domain code. One or more parameters of a machine learning based encoder are updated based on the one or more weights. Features are extracted from the one or more medical images using the machine learning based encoder with the one or more updated parameters. A medical imaging analysis task is performed based on the extracted features. Results of the medical imaging analysis task are output.

This invention relates to the field of image analysis technology, and more specifically, to a method for extracting and analyzing the dynamic range of images of damaged landforms in mining areas. The method includes: Step 1: Integrating the multispectral radiance within a set wavelength range, while subtracting two atmospheric scattering terms at each wavelength point to obtain the corrected reflectance; Step 2: Multiplying the Euclidean distance by the curvature amplitude in the near-infrared band to obtain a spectral collapse-curvature index sensitive to browning / chlorosis; Step 3: Calculating the peak elastic strain energy density per unit area of ​​overburden-coal mass; Step 4: Extracting all linear cracks from remote sensing images of the same time period and statistically analyzing their orientation histograms, then using the Shannon entropy formula to calculate the anisotropic entropy modulation of the cracks, thereby constructing a comprehensive dynamic range damage index to characterize the degree of damage to each pixel. This invention achieves accurate identification and intensity classification of surface damage areas under mining disturbance in mining areas.

The invention discloses an ultrasonic self-adaptive scanning method and system for EIT dynamic path planning, and belongs to the technical field of medical imaging. Reconstructing an impedance distribution image in real time based on EIT and detecting an abnormal region; the coordinates of the abnormal area are mapped to an ultrasonic coordinate system, and breathing phase compensation is carried out; planning an ultrasonic beam scanning path according to clinical constraint conditions; controlling the ultrasonic beam to scan along the planned path; performing image analysis on the scanning area; and updating system parameters according to an analysis result. According to the method, abnormal coordinates detected by EIT are accurately converted into an ultrasonic beam deflection angle, clinical constraint path planning of a high-cost area of a rib shielding area and breathing phase dynamic compensation taking end expiratory as a benchmark are combined, so that the system realizes full-closed-loop control from abnormal sensing, path planning to beam execution, and the accuracy of the system is improved. The problems of scanning failure caused by rib shielding and coordinate drifting caused by respiratory movement in clinic are effectively solved, and the reliability of dynamic monitoring of the thoracic cavity is remarkably improved.

The embodiment of the invention provides a pop-up window processing method and device and a storage medium. In the embodiment of the invention, based on a collaborative architecture of two AI models, pop-up window processing is divided into two stages of potential pop-up window area analysis and pop-up window feature recognition, a potential pop-up window area in a screen image is analyzed by using a first AI model, and a sub-image of the potential pop-up window area is cut out from the screen image; and performing multi-dimensional feature analysis and popup window feature identification based on the sub-images by using the second AI model, positioning position information of closing options according to the multi-dimensional feature analysis and the popup window feature identification, executing popup window closing operation based on the position information, cutting out the sub-images of potential popup window areas, converting global analysis into local analysis, reducing calculation overhead of irrelevant areas, and on the basis, improving the popup window closing efficiency. Model reasoning is used for analyzing the cut sub-images, the efficiency and accuracy of pop-up window detection and closing are effectively improved, then the probability that automatic testing is interrupted due to pop-up window interference is reduced, and the testing stability and continuity are improved.

Systems and methods for performing a medical imaging analysis task are provided. 1) One or more input medical images each in a domain and 2) a domain code for each of the one or more input medical images identifying its domain are received. Each particular input medical image of the one or more input medical images is downsampled using a dynamic convolutional layer based on the domain code for the particular input medical image. A first set of features is extracted from each particular downsampled input medical image of the one or more downsampled input medical images using the dynamic convolutional layer based on the domain code for the particular downsampled input medical image. The first set of features is encoded into a second set of features using a machine learning based encoder. A medical imaging analysis task is performed based on the second set of features using a machine learning based decoder. Results of the medical imaging analysis task are output.

The application discloses a radar chart-based ground foreign matter detection method, device, equipment and medium. The method comprises the following steps: acquiring a current radar image of a radar detection area; obtaining a current contour frame information library according to the current radar image; wherein the current contour frame information library comprises current contour frame information of at least one to-be-detected target; and performing foreign matter detection on the to-be-detected target according to the current contour frame information and a historical contour frame information library of the to-be-detected target; wherein the historical contour frame information library is determined according to the historical radar image. According to the technical scheme, accurate current contour frame information library can be obtained through analysis and processing of the current radar image, and then the current contour frame information of the to-be-detected target is compared with the historical contour frame information library, so that whether the to-be-detected target is a foreign matter can be accurately and quickly detected.

Embodiments of the present disclosure provide systems and methods for analyzing medical images containing vascular structures using a sequential model. An example system includes a communication interface configured to receive the medical image and the sequential model. The sequential model includes a vessel extraction sub-model and a lesion analysis sub-model. The vessel extraction sub-model and the lesion analysis sub-model are trained independently or jointly. The example system further includes at least one processor configured to apply the vessel extraction sub-model to the received medical image to extract location information of the vascular structure. The at least one processor also applies the lesion analysis sub-model to the received medical image and the location information extracted by the vessel extraction sub-model to obtain a lesion analysis result of the vascular structure. The at least one processor further outputs the lesion analysis result of the vascular structure.

A navigation system for a host vehicle is provided. The system may comprise at least one processing device programmed to receive, from a camera, a plurality of images representative of an environment of the host vehicle; analyze the plurality of images to identify at least one vehicle-induced occlusion zone in an environment of the host vehicle; and cause a navigational change for the host vehicle based, at least in part, on a size of a target vehicle that induces the identified occlusion zone.

This invention provides a rapid and intelligent assessment method for intestinal cleanliness based on colorimetric test strips, belonging to the field of medical testing technology. The method involves adding excrement samples to the test strip and acquiring time-series images, analyzing the spatiotemporal evolution characteristics of color development to construct a dynamic feature matrix, and then determining the composition and distribution of residues. By mapping hue migration trajectories to a color space and analyzing the geometric morphology and topological relationships of the temporal evolution path, a comprehensive cleanliness score is finally generated by coupling multi-dimensional features. This method achieves rapid, objective, and intelligent assessment of intestinal cleanliness.

The application relates to the technical field of water conservancy project safety monitoring and intelligent remote sensing image analysis, in particular to a water conservancy project abnormal deformation detection method and system; the method obtains multi-period remote sensing images of a drone, matches the SuperPoint and LightGlue models, and calculates deep feature distances to generate a difference map; the core is to adopt multi-modal large model preliminary screening and CLIP semantic alignment fine screening to eliminate false abnormalities and determine the final abnormal area; when an abnormality is detected, the system will automatically label, generate a report and output a warning; the application overcomes the non-precise alignment problem and realizes robust abnormality detection under complex shooting conditions.

A technique for determining the smearing of the kinetic energy of a hadron beam, using elements of the particle time of flight method, where all amplitudes for the measured signals SA(k) and SB(k) are recorded simultaneously from two detectors (A) and (B) located at a precisely defined distance L from each other and connected to the processing unit enabling the analysis of images recorded by both detectors, placed along the line of the studied hadron beam, over a period of time corresponding at least to 100 times the theoretical value for the time of flight of a hadron beam particle between the detectors (A) and (B), with time resolution at least on the level of 0.5 ns. Signals, for which the amplitude value is lower than the defined noise level and signals, the profile of which deviates from the determined reference profile, are eliminated.

The invention discloses a fishery enhancement and release operation monitoring method and system based on unmanned aerial vehicle cluster collaboration, and the method comprises the steps: dividing a plurality of monitoring regions according to a target water area, obtaining the high-definition image and multispectral data of each region through unmanned aerial vehicle cluster collaboration, and carrying out the secondary collaboration collection and correction of a fuzzy image; analyzing indexes such as water transparency based on the high-definition image to generate a first environment index, and analyzing parameters such as chlorophyll based on the multispectral data to generate a second environment index; weighting the environment index to obtain an environment estimation value, determining a multi-path migration area, and introducing a Dijkstra algorithm to analyze an environment moderate migration direction and a migration index; and setting a release quantity according to the migration condition, and generating a release planning scheme. The method solves the problems of low monitoring efficiency, one-sided cultivation release evaluation and the like in the traditional technology, improves the release scientificity and efficiency, and is suitable for a multi-water-area scene.

The invention discloses a navigation safety system for a mobile robot in an unknown environment, and belongs to the technical field of robot control. The navigation safety system comprises a searchlighting module used for acquiring searchlighting images, containing light beams, of the mobile robot in different directions; the path optimization module is used for analyzing dust sectional distribution conditions in different directions based on the lamplight searchlighting image and generating a navigation path optimization strategy; and the autonomous navigation module is used for executing basic moving path planning and dynamically adjusting navigation parameters in combination with a navigation route optimization strategy. According to the invention, based on the multidirectional light path image, region segmentation processing is carried out, the dust concentration of different sections is analyzed, exploration scenes of special environments such as collapse tunnels and pits are adapted, targeted technical support is provided for path decision of the mobile robot in a high-dust unknown region, and the navigation safety and task execution reliability of the mobile robot are improved.

The invention belongs to the technical field of intelligent operation and maintenance of power stations, and particularly relates to an intelligent operation and maintenance detection method, system and equipment for GIS equipment of a transformer substation, and the method comprises the steps: firstly collecting an infrared power station image and a multi-angle power station image of a target new energy power station, and segmenting an infrared assembly image from the infrared power station image; calculating the pixel temperature of the target transformer substation GIS equipment according to the infrared component image, screening out an abnormal area according to the pixel temperature, obtaining the temperature information of the abnormal area, collecting the power station electrical information of the abnormal area, and analyzing the power station component information of the target transformer substation GIS equipment based on the multi-angle power station image; and finally, inputting the power station component information, the power station electrical information and the temperature information into the fault diagnosis model, and outputting a fault diagnosis result. The input data features of the constructed model are more comprehensive, the fault detection precision can be effectively improved, only the data of the abnormal area of the power station needs to be processed during fault diagnosis, and the detection efficiency can be greatly improved.

The application relates to the technical field of medical treatment and artificial intelligence, in particular to an intelligent tongue diagnosis system and method, which comprises a camera transmission module, a server, a tongue image detection module, a tongue image segmentation module and a tongue image and constitution recognition module, and a client application; the camera transmission module is used for collecting original tongue images and transmitting the original tongue images to the server; the tongue image detection module is used for extracting tongue region images based on a deep learning target detection algorithm; the tongue image segmentation module is used for segmenting the tongue and the background based on a deep learning image segmentation algorithm to obtain tongue image images; the tongue image and constitution recognition module is used for performing feature extraction and analysis based on a deep convolutional neural network and outputting tongue image feature recognition results and traditional Chinese medicine constitution classification results; and the client application is used for displaying the results; thus, through separation of image collection, server-side intelligent analysis and client-side display, professional and efficient image analysis is realized while the portability of equipment is ensured.

The invention discloses a cataract ultrasonic emulsification postoperative cornea thickness detection method based on anterior segment OCT, and relates to the technical field of image analysis. The method comprises the following steps: dividing a cornea into a plurality of first regions based on a region growing algorithm and initial cornea thickness data; acquiring a second area based on each first area; acquiring a plurality of partial boundaries based on the second region; based on the boundaries of all the parts, a fibrosis area is obtained; and adjusting the scanning density and the refractive index of the second region based on the fibrosis region and the scanning density correction coefficient to finally obtain cornea thickness data. According to the method, the scanning density and the refractive index of the second area are adjusted by combining the fibrosis area and the scanning density correction coefficient, so that the error of calculation by using a unified refractive index in the prior art is avoided, accurate measurement of the cornea thickness is finally realized, and the technical problem that the thickness of the cornea area cannot be accurately measured through existing OCT detection is solved.

The present invention provides a full-stack medical imaging foundation model, re for performing medical image segmentation, classification and localization using a unified language-image correlation mechanism. The present invention leverages pretrained language-guided associations to automatically recognize and localize disease targets without requiring medical imaging knowledge or manual box-prompting. The invention enables comprehensive clinical tasks across multiple disease classes and various medical imaging techniques, improves segmentation efficiency, and reduces human input burden. With operability through human language-only prompts by non-radiology specialists, the present invention improves real-world applicability and enables deployment of foundation model-based segmentation in diverse clinical environments.

The invention discloses a left ventricular ejection fraction regression prediction system based on transfer learning and multi-view fusion, and belongs to the field of medical image analysis. The problem that an existing deep learning method is poor in left ventricular ejection fraction prediction accuracy is solved. According to the method, the A4C view angle echocardiogram data and the PSAX view angle echocardiogram data are subjected to independent model pre-training, and then the pre-trained model parameters are subjected to transfer learning, so that a better initialization effect can be obtained, and model convergence and performance optimization are accelerated. The double-view-angle fusion regression module can well combine spatial and temporal features output by the two single-view-angle models, feature relevance of echocardiogram data of the two view angles is greatly utilized, and the accuracy of final model prediction can be improved. In addition, through fusion of the multi-view echocardiogram data features, the influence of anatomical structure differences on the detection result is effectively reduced. The method can be applied to left ventricular ejection fraction regression prediction.

The invention belongs to the technical field of environment monitoring, and discloses a forestry ecological environment monitoring system and method, and the system comprises a plant monitoring module which is used for monitoring a plant key; the meteorological data acquisition module acquires meteorological data in real time; the route planning module plans an optimal flight route; the cruise monitoring module carries out all-directional cruise monitoring; the image analysis unit analyzes and recognizes the collected image; the infrared data analysis unit analyzes and identifies the infrared data to identify abnormal conditions; the remote sensing data analysis unit analyzes the remote sensing data and identifies abnormal conditions; and the comprehensive analysis module fuses the multi-source data, comprehensively analyzes the multi-source data and predicts potential risks. According to the invention, omnibearing dead-corner-free coverage is realized through monitoring dimensionality diversification, unmanned aerial vehicle cruising is combined with satellite remote sensing, and monitoring from the air to the macroscopic level is realized; and the comprehensive analysis module fuses multiple types of data, performs comprehensive analysis, predicts potential risks, timely performs risk early warning, and actively intervenes to reduce disaster loss.

The system according to this embodiment aims to analyze information on a child's symptoms and injuries and provide information to help parents take appropriate action. [Solution] The system according to the embodiment comprises a reception unit, an analysis unit, a sharing unit, and a notification unit. The reception unit inputs images of the child's symptoms or injuries. The analysis unit analyzes the information input by the reception unit and presents possible diagnoses, whether hospital visits are necessary, and the estimated number of days until complete recovery. The sharing unit shares the information obtained by the analysis unit with other parents. The notification unit sends a push notification to check on the progress based on the information shared by the sharing unit.

This invention provides a full-stack medical imaging foundation model for performing medical image segmentation, classification, and localization using a unified language-image correlation mechanism. This invention leverages pre-trained language-guided associations to automatically identify and localize disease targets without requiring medical imaging knowledge or manual bounding box prompts. This invention enables comprehensive clinical tasks across multiple disease categories and various medical imaging techniques, improving segmentation efficiency and reducing the burden of manual input. By utilizing the operability of plain text prompts from non-radiologists, this invention improves real-world applicability and enables the deployment of foundation model-based segmentation in diverse clinical settings.

This invention discloses a method and system for unmanned aerial vehicle (UAV) forest patrol based on BeiDou positioning, relating to the field of forest patrol. It includes: a region limitation module for uploading location coordinates and limiting the patrol area based on the uploaded coordinates; a control module for controlling the UAV to fly within the patrol area and collect forest images; and an analysis module for receiving the forest images collected during the UAV's flight, storing the images, and analyzing the forest health trend based on the stored images. This invention, through the design and deployment of the limited patrol area and the UAV analysis logic, enables the UAV to autonomously and intelligently complete the collection of forest images. Furthermore, during the image collection process, the forest image acquisition path is autonomously and randomly generated, ensuring the diversity of forest image collection and guaranteeing that the subsequent assessment of the forest ecological environment's health by the system is more reliable and effective.

The system according to the embodiment aims to efficiently record and organize daily conversations and events. [Solution] A system according to an embodiment includes a recording unit, an analysis unit, a classification unit, and a generation unit. The recording unit records conversations and images. The analysis unit analyzes the data recorded by the recording unit. The classification unit separates the data analyzed by the analysis unit into categories. The generation unit creates a diary based on the data classified by the classification unit.

The present disclosure provides a determination method of a scanning protocol, a control method of a medical imaging device and a system thereof. The determination method comprises: inputting medical record information into a first protocol matching model and outputting a first target protocol package; the first target protocol package is used for preliminary scanning to generate a first scan image; in response to the presence of a region of interest related to a medical event in the first scan image, the region of interest is automatically analyzed to obtain a qualitative analysis result or a quantitative analysis result of the medical event; and in response to the qualitative analysis result or the quantitative analysis result of the medical event not meeting a set requirement, the first scan image is input into a second protocol matching model, and it is determined whether a second target protocol package is output. The automatic medical record information processing and image analysis, intelligent recommendation and supplementary scanning protocol are realized, the diagnosis and treatment scanning efficiency is significantly improved, unnecessary repeated examinations are reduced, the diagnostic accuracy is enhanced, and the patient experience and medical resource allocation are optimized.

The invention discloses a machine vision-based warehouse cargo automatic sorting manipulator, which comprises a fixed base, a plurality of mechanical arms, a plurality of mechanical arms, a plurality of mechanical arms and a plurality of mechanical arms, wherein the fixed base is mounted on one side of a cargo conveying frame; the steering seat is rotationally assembled on the upper end surface of the fixed base; one end of the main machine arm is rotationally connected to the steering seat; the auxiliary machine arm is rotationally connected to the other end of the main machine arm, and a rotating arm is coaxially and rotationally mounted on the auxiliary machine arm; the two clamping units are arranged in parallel; the visual sensor is mounted in the middle of the lower end face of the rack; the front scanning device can conduct all-dimensional scanning on the goods in advance, so that the shape specification and the optimal clamping direction of the goods are analyzed based on scanned images, and a clamping unit mainly arranged on the sorting mechanical arm can absorb and fix the surfaces of the goods through a suction cup unit and can also absorb and fix the surfaces of the goods through the suction cup unit. And the flexible locking assembly can be used for being in contact with the surface of the goods for clamping, and the adaptability and universality of the sorting manipulator are greatly improved.

The invention provides a dynamic light source adjusting method and system based on environmental perception and a medium. The method comprises the following steps: setting a sampling frequency, collecting environmental illumination intensity and spectral data, and collecting a target image based on an industrial camera; constructing a reflection modeling network, and performing reflectivity analysis on the measured object based on the reflection modeling network to obtain reflection characteristics of the surface of the measured object to different wavelengths; outputting a lighting strategy based on the lighting strategy network; generating an optimal illumination spectrum based on an illumination strategy, and outputting according to the optimal illumination spectrum based on a driving LED array to obtain a light source environment; a new measured object image is collected based on the light source environment, the contrast ratio of the measured object image is analyzed, and the environment illumination intensity and wavelength distribution are dynamically adjusted based on a set contrast ratio threshold value; the intensity and wavelength distribution of an illumination source are dynamically adjusted by sensing the ambient light intensity and spectral distribution in real time in combination with deep modeling of the reflection characteristics of the surface of a measured object, and the maximum imaging contrast and recognition effect are achieved.