14 results about "Computer vision algorithms" patented technology

The present application belongs to the technical field of oil use peony phenotype parameter automatic extraction method, especially relates to a kind of oil use peony phenotype parameter automatic extraction method and system based on multi-organ image analysis, obtain the high-resolution original image of oil use peony multi-organ containing ColorChecker24 standard color card, the organ at least includes petal, blade, fruit pod;Color correction and geometric correction are executed to original image, 24 color blocks of color card are positioned using computer vision algorithm, color conversion matrix is constructed to map image from device-dependent RGB space to standard color space, while calculating pixel scale based on color block physical size;Four corner points of bearing tray are detected for fruit pod image, and perspective transformation matrix is constructed to complete perspective distortion correction.

The application provides a real-time conversion method for pixels and geographic latitude and longitude in low-altitude photogrammetry, which comprises the following three parts: unmanned aerial vehicle capability confirmation, real-time conversion service implementation deployment for pixels and geographic latitude and longitude in low-altitude photogrammetry, and GIS geographic information platform integration docking. Through real-time data collected by an unmanned aerial vehicle, combined with a computer vision algorithm and a spatial geometric transformation model, the specific pixel points in image data are associated with real geographic coordinates, so that accurate positioning and measurement of ground objects are realized. The application relates to the technical field of low-altitude photogrammetry, and through the cooperative work of real-time data collection by an unmanned aerial vehicle, real-time conversion service for pixels and geographic latitude and longitude in low-altitude photogrammetry, and a GIS geographic information platform, efficient mapping from a photographed image to a geographic coordinate is realized, and accurate and rapid technical support is provided for low-altitude photogrammetry.

This invention discloses a navigation system based on an intelligent bicycle, belonging to the field of navigation technology. It addresses the problems of path planning and safety control under dynamic road conditions. The system first constructs a digital situational awareness model of the vehicle by synchronously acquiring data from a multi-dimensional motion sensor group and wheel speed pulse sequences to calculate real-time attitude angles, speed decay, and lateral sway amplitude. A road condition analysis module uses computer vision algorithms to reconstruct a three-dimensional digital model of the road surface, identifying height abrupt changes and determining obstacle crossing costs. A cost fusion module uses a multi-modal data fusion algorithm to map vehicle condition parameters into energy factors and generates a navigation cost chain that has undergone local smoothing. A path generation module performs path search based on this cost chain and outputs the optimal steering command. This invention achieves a deep mapping between the physical driving environment and the digital cost space, significantly improving navigation accuracy under complex dynamic road conditions.

The application discloses an intelligent interactive projection method and an interactive projection device, and relates to the technical field of interactive projection.The application solves the technical problem that the recognition accuracy is low, the algorithm cannot accurately detect the key features of user actions, and the interaction is affected.The application generates signals by detecting the detailed features of user gestures and the multi-dimensional information of specific objects in the motion signal generation stage for static motion signals.Compared with some simple motion detection methods, the application can more accurately recognize static actions.For real-time motion signals, the application improves the accuracy of real-time motion recognition by comprehensively capturing motion features, and efficiently processes image data by using advanced computer vision algorithms and signal processing technologies.In the real-time motion signal processing process, the application meets the real-time requirements by quickly preprocessing images, extracting features and processing in segments, and guarantees the smoothness of the interaction.

The present invention discloses a system to assist investigations for storage and targeted search of metadata and video collected by dashcams. The invention efficiently guides video evidence search using stored route metadata versus complete footage, tackling the challenges of distributed footage across vehicles. The system enhances investigations by unlocking previously inaccessible visual evidence in a targeted manner while optimizing storage and bandwidth. Vehicles equipped with cameras send real-time route metadata and / or video to the system. This data may include coordinates, timestamps, license plates detected, facial features, street names, addresses, suspicious behavior data, and other details related to the route. Investigators may input search parameters related to location, timing, vehicles, objects, persons, etc. Computer vision algorithms analyze stored metadata to identify matching routes and footage. Investigators may access the video from central storage or, if only metadata about the route was submitted, request relevant footage from the respective driver.

FIELD: video surveillance systems.SUBSTANCE: invention can be used in territorially distributed automated security systems. Initial data is pre-formed and specified: a list of target objects, reference images of target objects, the target object image match coefficient, the background image change coefficient, the initial background image update time period, and video camera parameters in the video monitoring tool. Video data is received from a digital video camera and a thermal imager and the video is split into a sequence of frames. The first frame is recorded into the storage device and sent to the database server as a background image, which is updated according to the specified time period. From the second frame, computer vision algorithms are used to search for target objects, for which each new frame is sequentially compared with the saved background image until the end of the video surveillance period.EFFECT: reducing the volume of video data traffic transmitted over communication channels while maintaining the image quality of specified objects and increasing the security of the transmitted information.5 cl, 6 dwg

Provided is an unmanned radish harvesting system, relating to the field of agricultural artificial intelligence technologies. The system, using an unmanned radish harvesting control subsystem, an unmanned radish harvesting operation subsystem, and a radish agronomic parameter configuration subsystem, allows for intelligent control of the movement of a radish harvester and optimal operation path planning by integrating the shape and terrain of a vegetable field plot and planting agronomic practices. On the basis of various parameter configurations, a computer vision algorithm is employed to accurately identify radish fruits and radish tops. The unmanned radish harvesting system realizes more intelligent, delicate, and accurate operations of the unmanned radish harvester, thereby saving human resources, material resources, and time, and improving the efficiency and quality of unmanned radish harvesting as well as the green and intelligent levels of unmanned radish harvesting.

This invention discloses an intelligent sales strategy optimization system and method based on multi-source data, belonging to the field of sales strategy optimization technology. The invention acquires and preprocesses outfit data from mainstream video platforms, extracts visual style features using computer vision algorithms, establishes association mappings by combining outfit text style keywords, constructs a style feature system, and categorizes outfits to form popular style clusters. It then digitally models in-store clothing, aligns with the style feature system dimensions to build an outfit library, and uses quantitative features of popular clusters as conditions to retrieve and filter outfits with matching scores through similarity algorithms, generating an associated list and targeting it across multiple channels. The invention also statistically analyzes outfit clicks and purchases during the promotion period, calculates the click-to-purchase ratio, constructs an outfit association topology, and analyzes its features; finally, it dynamically adjusts inventory configuration based on the click-to-purchase ratio, topology features, and existing inventory.

This invention discloses a physical education examination system for assessing student academic performance, belonging to the field of physical education examination systems. It includes: an image acquisition module equipped with multiple high-definition cameras arranged around a throwing circle to collect video data of the examinee during the throwing process; an identity verification module for performing two-stage identity verification; a foul detection module for analyzing video data to detect various fouls; a score processing module for handling score disputes; when there is an objection to the cancellation of a score due to a foul, the system retrieves the surveillance video, extracts a segment of the foul, enlarges it, and displays it on a large screen in the examination area to ensure fairness and transparency; and an output module for providing examinees with a valid score report after the examination, and separately listing areas for improvement. This invention, by using computer vision algorithms to replace or assist human judgment, eliminates the subjective bias of referees, making the identification of fouls more accurate and consistent, and greatly ensuring the fairness of the examination.

A computer-implemented method, a computer system and a computer program product recognize relevant objects in images in real time. The method includes capturing an image of a surrounding environment using a device. The method also includes identifying an object and a shape of the object in the image of the surrounding environment using a computer vision algorithm. The method further includes determining that the object is a relevant object based on the shape of the object. Lastly, the method includes performing an operational action based on the relevant object.

The application discloses an ultrasonic uterine junction zone evaluation system, and the subjective process is converted into objective mathematical calculation through image processing and computer vision algorithm. Specifically, after the image quality is optimized by using anisotropic diffusion filtering and CLAHE algorithm, a hybrid segmentation technology based on edge detection and three-dimensional region growth is adopted to accurately extract a three-dimensional geometric model of JZ. On this basis, the integrity of JZ is quantified by defining and calculating a core parameter, namely, continuity index (CI). The algorithm samples along the center line system of JZ, analyzes the mutation of the local width of adjacent points, and accurately represents the continuity degree of the ring structure by using a mathematical proportion (CI value).

A system for monitoring shopping baskets (e.g., baskets on human-propelled carts, motorized carts, or hand-carried baskets) can include a computer vision unit that can image a surveillance region (e.g., an exit to a store), determine whether a basket is empty or loaded with merchandise, and assess a potential for theft of the merchandise. The computer vision unit can include a camera and an image processor programmed to execute a computer vision algorithm to identify shopping baskets and determine a load status of the basket. The computer vision algorithm can comprise a neural network. The system can identify an at least partially loaded shopping basket that is exiting the store, without indicia of having paid for the merchandise, and execute an anti-theft action, e.g., actuating an alarm, notifying store personnel, activating a store surveillance system, activating an anti-theft device associated with the basket (e.g., a locking shopping cart wheel), etc.

A method of processing agricultural images includes comparing object detections performed by multiple image processing schemes to determine a set of ground truth images from which at least one machine learning (ML) models used by at least one ML algorithm included in the multiple image processing schemes is trained, wherein the multiple image processing schemes include two or more of (a) an image processing scheme that includes a cascade of multiple ML algorithms; (b) an image processing scheme that includes image annotation based on user feedback; or (c) an image processing scheme that includes a cascade of an ML algorithm or a computer vision (CV) algorithm and a user feedback.

A system for the automated detection and management of traffic violations, comprising: a surveillance camera network with a multitude of cameras, each camera configured to record continuous video streams for monitoring vehicle movements and potential traffic violations, and processing the recorded video streams for license plate and traffic violation recognition; a license plate recognition module communicatively connected to the surveillance camera network and configured to receive the continuous video streams, the license plate recognition module further configured to extract and process license plate images from the video stream to accurately identify offending vehicles and to determine the vehicle registration data from the extracted license plate images;a machine learning processing unit communicatively connected to the surveillance camera network, configured to receive the continuous video streams, wherein the machine learning processing unit is further configured to analyze the video streams using computer vision algorithms and detect traffic violations such as speeding, running red lights, illegal parking and accidents; a real-time data processing module configured to: receive violation data from the machine learning processing unit and vehicle identification data from the license plate recognition module, correlate the received data with learned violation patterns, generate violation records containing timestamps, location and vehicle details, and transmit the violation records to law enforcement agencies and users;a user notification and resolution interface configured to send automated notifications of violations to vehicle owners via SMS, email, or mobile application, the user notification and resolution interface further configured to provide an interface for reviewing evidence of violations, contesting violations, and settling penalties; an emergency services integration module configured to detect accidents and, upon detection, automatically notify nearby hospitals and police stations; and a traffic management integration interface that interacts with existing traffic infrastructure and APIs to provide parking advice, manage violation workflows, and support community-reported incidents.