16 results about "Automated algorithm" patented technology

Disclosed in the present invention are an edge-deployed semi-supervised anomaly detection method and system for a railway track foreign object. The method comprises the following steps: an edge device encoding and decoding a video stream captured by a camera to obtain an image frame sequence, and performing frame extraction; and using a semantic segmentation model to perform image segmentation on a certain image frame obtained by means of frame extraction, to obtain a railway track region segmentation image. The use of a single image as input may generate an expert model result having a high weight value; however, the determination based on a single image is not stable, multiple consecutive images of the task scene need to be inputted, the frequency of each expert model obtaining the highest weight is computed, and the expert model corresponding to the highest frequency is the final solution. The present invention supports scene-adaptive foreign object detection algorithm automatic selection, and a user can perform selection on the basis of prior knowledge, or selection may be performed by a scene-adaptive automatic algorithm selection method; the user only needs to provide a batch of image data of the current scene, and the optimal algorithm selection can be evaluated.

The application discloses an ancient landslide dam extraction method based on optical images, and specifically comprises the following steps: step one, image acquisition; step two, basic element indication; step three, ancient landslide interpretation; step four, river blocking analysis; step five, dam interpretation; step six, parameter calculation and verification; and step seven, database construction. The application relates to the technical fields of remote sensing interpretation and ancient disaster identification. The ancient landslide dam extraction method based on optical images combines the ancient landslide range outlined on the images with direct interpretation signs of ancient dams and indirect inference methods, can preferably extract the ancient dams, and is supplemented by field sampling verification. Compared with an automatic algorithm identification mode, the ancient landslide dam extraction correctness is greatly improved, and the time and labor of pure field investigation are effectively avoided, and the efficiency is greatly improved.

This invention belongs to the field of dental medical technology, specifically relating to an automatic positioning method, electronic device, and storage medium for dental template teeth. The method includes: acquiring a patient's dental arch model, the cervical margin of a restoration, and a template tooth model; determining the dimensional characteristics of the template tooth; performing preliminary overall matching of the template tooth; extracting the tooth portions from the dental arch and fitting a dental arch curve; calibrating the position and size of the template tooth based on the dental arch curve, including determining the target position and orientation of the restoration based on the cervical margin and the dental arch curve for restored teeth, and determining and locating the missing tooth based on the projection of adjacent restorations, and scaling it according to the morphology of the cervical margin or the arc length of the missing tooth interval. This invention achieves precise initial positioning and adaptation of the template tooth through an automated algorithm, solving the problems of cumbersome and inefficient manual positioning in existing technologies.

The invention relates to the technical field of intelligent maintenance and artificial intelligence crossing of civil aviation, and provides a civil aircraft health management method and device based on a large model agent. According to the method, through combination of natural language interaction and an automatic algorithm, the workload of glue code writing and parameter debugging is reduced, the PHM system development threshold is lowered, and non-algorithm professional engineers can quickly construct an analysis process; task planning and algorithm calling are autonomously completed, the process execution efficiency is optimized, and the task processing period is shortened.

The invention belongs to the technical field of underwater acoustic signal processing, and particularly relates to an active sonar man-machine cooperation multi-target tracking method. Comprising the following steps: carrying out balanced denoising processing on background noise, and automatically detecting a threshold value to obtain azimuth and distance information of a target; meanwhile, the detection threshold value is manually adjusted through manual intervention; establishing a track for the detection point through an automatic algorithm and distributing a track ID; meanwhile, through manual intervention, suspected target points are manually recorded; calculating the proximity of each detection point and each track through an automatic algorithm, and carrying out track association according to a proximity matrix; meanwhile, through manual intervention, track ID affiliation is manually corrected; according to the invention, a hybrid tracking mode of'automatic-based and manual-assisted 'is formed, a full-automatic tracking algorithm is used for processing a conventional process, an abnormal decision is manually intervened, a high-reliability man-machine cooperative active sonar target tracking system is established, and the tracking robustness in a complex environment is improved by dynamically switching a manual / automatic mode.

The application relates to the technical field of robots and discloses a method for remotely manually repositioning coordinates of an indoor robot, which comprises the following steps: step 1, identifying a positioning loss state, and the platform sends an action: when the robot cannot maintain reliable positioning, the platform timely sends an alarm, and after an operator receives and confirms the alarm, the repositioning process is started; step 2, constructing a new map: a new map construction mechanism is set; step 3, calculating a target pose: the operator manually aligns the new map and the old map to obtain the target pose; and step 4, repositioning: the robot realizes internal updating according to the target pose. The application combines intelligent guidance of an automatic algorithm and flexible decision-making of manual operation, effectively overcomes the defects that pure automatic repositioning is prone to failure in a large range loss or a complex environment and falls into a local optimum, and simultaneously avoids the problems that traditional manual intervention needs on-site operation and is low in efficiency.

Mechanisms are provided for automatically determining a speaker or writer's viewpoint, intensity, veracity and intended audience of natural language text. Language understanding functions including question answering, information search and retrieval, text generation, machine translation, transcription, summarization and decision-support are more accurate when these indirectly articulated elements of intent are understood. The functions use explicitly coded and curated commonsense knowledge propositions that form a knowledge graph data structure to interpret these elements of intent using contextual, subtextual and extratextual cues. These mechanisms operate on input text that has already undergone lexical disambiguation and provide a framework for identifying bias, deception and literary devices including simile, metaphor, analogy, allegory, personification, hyperbole, metonymy, sarcasm, diplomacy, intimidation and allusion.

The invention discloses an EN50399 standard-oriented flame burning area envelope line digital detection method and device. The method comprises the following steps of: 1, reading and preprocessing an image; step 2, contour detection and light-emitting plate rectangle screening; step 3, quadrant division and target confirmation; 4, performing perspective transformation and image splicing; step 5, low R channel square positioning is carried out to establish a real coordinate system; 6, carrying out dark area contour fitting and feature extraction; the device comprises three modules, namely a core control module, an image acquisition module and a calibration reference module. According to the method, through standardized hardware calibration and automatic algorithm processing, manual naked eye recognition and manual marking are replaced, the problem of judgment threshold difference caused by personnel experience in a traditional method is solved, and the reproducibility of a detection result is high; and moreover, expensive equipment is not needed, the portability is good, and the method is suitable for field detection scenes of combustion equipment production enterprises, quality inspection institutions and certification laboratories.

The application provides a method and device for identifying dead farmed animal individuals in a video based on a ViT technology, a storage medium and a terminal, and belongs to the technical field of computers. The method is applied to a terminal. The terminal selects key frames from a short video segment containing dead farmed animals, and synthesizes pseudo-color three-channel images from the key frames. The pseudo-color three-channel images are data-labeled to label dead farmed animal individuals and non-dead farmed animal individuals in the images, and a training sample set is obtained. The training sample set is input into a ViT target detection model to train the model. The scheme converts target detection in a video into target detection in an image, thereby reducing the time cost of labeling work. Whether the farmed animals are alive or not is inferred by distinguishing the motion and stillness of the animals in the short video segment. When there are tens of thousands or even millions of animals in a large-scale farm, the automatic algorithm can replace the frequent patrol cost of manual work.

Systems and methods are disclosed for automatically performing motion correction in dynamic positron emission tomography scans, such as dynamic positron emission tomography myocardial perfusion imaging studies. An automated algorithm can be used. The algorithm can use simplex iterative optimization of a count-based cost-function customized to different dynamic phases for performing frame-by-frame motion correction.

The application discloses a kind of flame scorching area envelope digital detection method and device for EN50399 standard, method includes: step 1, image reading and pre-processing;Step 2, contour detection and light-emitting plate rectangular screening;Step 3, quadrant division and target confirmation;Step 4, perspective transformation and image splicing;Step 5, low R channel square positioning establishes real coordinate system;Step 6, dark area contour fitting and feature extraction;Device includes three modules, respectively: core control module, image acquisition module and calibration reference module.The application replaces artificial naked-eye identification and manual marking by standardized hardware calibration and automated algorithm processing, solves the problem of determination threshold difference caused by personnel experience in traditional method, and the detection result has strong reproducibility;And without expensive equipment, good portability, applicable to the on-site detection scene of combustion equipment production enterprise, quality inspection organization and certification laboratory.

This invention discloses a non-contact cardiac 3D mapping method and system, belonging to the field of image mapping technology. The method involves acquiring a sequence of cardiac ultrasound images across multiple cardiac cycles; preprocessing each cardiac ultrasound image to obtain optimized cardiac ultrasound images, including image enhancement and edge detection; registering each optimized cardiac ultrasound image based on a geometric constraint matching algorithm to ensure they are in the same coordinate system; performing 3D reconstruction using a triangulation algorithm based on the registered optimized cardiac ultrasound images to obtain a 3D cardiac model; performing 3D mapping of the cardiac structure on the 3D cardiac model; and outputting the 3D cardiac mapping results. The use of automated algorithms to process cardiac ultrasound images generates a high-quality 3D cardiac model for mapping in a shorter time, improving operational efficiency and accuracy compared to traditional manual operations, avoiding the subjective influence of manual mapping, and reducing surgical safety risks.

Method for producing a thin-walled structure from a starting material (1) by additive manufacturing in which a continuous, uninterrupted contour of the thin-walled structure to be manufactured is defined in a control unit (5), wherein The contour to be produced is divided into different geometry classes, and for each of the different geometry classes, the exposure strategies to be used and their parameterization are then determined manually or automatically using an algorithm. In a manufacturing process step, a radiation source (3) is controlled by the control unit (5) such that the defined contour is traced by an energy beam (4) emitted by the radiation source (3), at least one layer of the starting material (1) is irradiated along the contour and is thereby locally melted or sintered with the energy beam (4), so that the thin-walled structure is formed by an energy input, wherein the uninterrupted continuous contour is divided into at least two sub-sections, each of which is at least partially traversed by a single-line exposure, and wherein spatially directly adjacent subsections are irradiated with the energy beam (4) in such a way that an endpoint of one subsection irradiated by single-line exposure is not used as the starting point of the other subsection to be irradiated by single-line exposure in immediate succession.

This invention discloses a method and system for optimizing the layout of optic nerve electrode arrays. The method first constructs a static physiological model of the random distribution of optic nerve fibers. Then, it uses multiphysics simulation software to establish optic nerve and electrode models to obtain the potential distribution. A Python master script initializes the covariance matrix adaptive evolution strategy algorithm and introduces constraints related to biosafety and electrode manufacturing processes. Automated coupled modeling is achieved through a Python interface, mapping the potential distribution to an axon model for neurodynamic simulation and fitness evaluation. Finally, the algorithm iteratively updates and outputs the globally optimal set of electrode spatial distribution parameters. This invention deeply integrates three-dimensional finite element electric field simulation with neurodynamic models, and through automated algorithms, globally optimizes under strict constraints, resolving the contradiction between "high coverage" and "low damage" in optic nerve prosthesis design, significantly improving the electrical stimulation coverage and design efficiency of optic nerve prostheses.

Methods for the development and integration of multiple apparatuses and methods for achieving administration of stem cell therapies include precision manufacturing of tissue scaffolds and / or bioreactor substrates. The nano / microscale fiber material extrusion typifying the electrospinning process is married with the fiber alignment and layering characteristic of an additive manufacturing process. The method generates porous fibrous 3-D meshes with precision controlled structures from biopolymer melts and solutions and gels, blends, and suspensions with and without cells. A method of tracking the migration histories and shapes of stem cells on scaffold surfaces relies on immunofluorescent imaging and automated algorithms based on machine learning. The combination of the precision manufacturing method and the method of cell tracking and cell shape statistics, along with understanding of the intimate relationship between the cell shape / phenotype and scaffold architecture leads to an integrated method for cultivating and harvesting cells having desired phenotypes.

The present application belongs to the field of tunnel and underground advanced geological prediction, and provides a real-time detection method and system for unfavorable geology based on a drill-while-drilling seismic source of a rock drilling jumbo, comprising the following steps: S1: synchronous acquisition of the drill-while-drilling seismic source signal and the sensing signal; S2: strong background noise suppression and signal preprocessing; S3: seismic source signal extraction and wave field separation; S4: real-time imaging of the geological reflection interface and identification of unfavorable geology; and S5: information integration and real-time early warning. The present application establishes a rapid and automated algorithm process from signal acquisition, preprocessing, wave field separation, reflection wave extraction to geological interpretation, and gives a judgment while the construction is in progress, so that the construction is not disturbed, the efficiency is greatly improved, the time and economic cost of detection are reduced, and the construction safety is improved while the detection is performed while drilling, thereby providing a real-time detection method for unfavorable geology based on a drill-while-drilling seismic source of a rock drilling jumbo for safe construction of underground projects.