17results about How to "Avoid image distortion" patented technology

This application provides an image processing method, apparatus, computer-readable medium, and computer device. The image processing method includes: acquiring multiple original images and descriptive text corresponding to each of the multiple original images; converting the descriptive text corresponding to each of the multiple original images into text vectors to obtain text vectors corresponding to each of the multiple original images; performing clustering processing based on the text vectors corresponding to each of the multiple original images to obtain multiple original image clusters, and generating text embedding vectors corresponding to each original image cluster; using the text embedding vectors corresponding to each original image cluster as text-guided features, performing semantic editing processing on the original images in each original image cluster to obtain processed images, wherein the processed images are used as sample images for a machine learning model. The technical solution of this application embodiment can enrich the diversity of training datasets, thereby improving the generalization ability and robustness of the model.

The application discloses a burr detection mechanism of a battery pole piece, which comprises a support, a linear motion unit and a detection assembly, wherein the linear motion unit is installed on the supporting surface of the support; the detection assembly comprises a mounting frame, a fixing frame, a camera and a light guide piece, the camera is located above the light guide piece, and the optical axis of the camera is obliquely arranged relative to the light emitting surface of the light guide piece, so that the light of the pole piece to be detected is guided to the camera through the light guide piece; the mounting frame is connected with the linear motion unit; the fixing frame comprises a first fixing plate and a second fixing plate; the first fixing plate is connected with the mounting frame; the camera is fixed on the second fixing plate; a locking hole is formed in the side of the second fixing plate close to the first fixing plate; an arc-shaped opening is formed in the first fixing plate; and the opening is matched with the locking hole. Compared with the traditional manual microscopic sampling inspection mode, the burr detection mechanism can avoid the blind area of static detection, is suitable for dynamic detection on a continuous production line, and is favorable for improving the detection efficiency and the detection quality.

The utility model discloses a differential rudder wheel with code reading camera relates to the technical field of latent tractor assembly, including drive wheel, slewing support, motor, parallel shaft speed reducer, articulated axle, mounting flange and camera module, drive wheel is by motor through parallel shaft speed reducer drive, slewing support is connected with latent tractor car body through mounting flane, and mounting flange realizes the left and right tilt of rudder wheel whole through articulated axle, and the motor and parallel shaft speed reducer of symmetrical design enclose camera module in rudder wheel bottom center position, and the motor and parallel shaft speed reducer are higher than camera module and are used for protecting camera module. The utility model is to solve the technical problem in at providing a differential rudder wheel with code reading camera to solve the technical problem that latent tractor is deficient in steering flexibility, compact structure and visual navigation precision.

This invention relates to the field of remote sensing technology and discloses an automated collaborative verification method and system for highway slopes using UAVs based on ground-sensor event-driven methods. The method includes: acquiring ground sensor monitoring data and basic geographic information data of the slope; triggering alarm events by threshold judgment of the ground sensor monitoring data; generating regions of interest for the slope based on the basic geographic information data of the slope and matching modeling parameters to obtain a terrain-simulating modeling flight path; calculating the effective geometric parameters and average normal vector of the slope surface based on the three-dimensional coordinates of the sensors that triggered the alarm, constructing a local coordinate system for the slope surface, and obtaining a gridded emergency inspection flight path; encapsulating the terrain-simulating modeling flight path and the gridded emergency inspection flight path, and distributing the flight path tasks to the UAV hangar through a cloud application programming interface; the UAV executing the flight path tasks and transmitting back inspection data, performing multi-dimensional analysis on the inspection data, and completing a ground-air linkage closed-loop decision-making process.

The invention discloses omnibearing CT scanning equipment, and relates to the technical field of medical equipment. The body position of a patient can be accurately adjusted. According to the equipment, stable high-speed fluid is generated by driving the paddles, and the supporting table top and the lying plate are stably lifted and inclined through the lifting wings and the balance mechanism. The inclination angle control electric cylinder and the inclination angle sensor form closed-loop control, so that the accuracy of the inclination angle is ensured. And through the design of a stable supporting structure and a rolling assembly, the stability and safety of the equipment in the movement process are enhanced. According to the equipment, the problems of scanning blind areas, mechanical instability and the like existing in traditional CT scanning are solved. By monitoring the flow rate, the inclination angle and the position of the fluid in real time, the system can adjust operation parameters in a self-adaptive mode to meet the requirements of different patients.

This invention relates to a high-precision visual inspection device for pet food production lines, belonging to the field of pet food inspection technology. It includes a main body with a conveyor belt installed inside. A first inspection component is located at the upper end of the conveyor belt, used for preliminary inspection of stains and damaged areas on the material surface. A centering and rejection component is located on one side of the first inspection component, used for centering and rejecting unqualified materials during the conveying process. A first damage detection component is located inside the centering and rejection component, used for initial verification of damaged materials detected by the first inspection component. This invention significantly reduces the rate of missed and false detections through a tiered inspection mode of initial inspection with an industrial camera, initial verification with heating thermal imaging, and secondary verification with airflow temperature difference.

The invention discloses a verification method and device for a double-view lens, and the method comprises the steps: taking a circular array with the same diameter in the center of a large-view lens as a central circular array, and carrying out the distortion verification of the large-view lens based on the central circular array and a preset standard distance; the standard circles are arranged at the four quadrant positions and the center position of the standard lens respectively, and measurement information of the corresponding positions of the standard circles is obtained respectively; performing pixel verification of the standard lens based on the measurement information of the corresponding position; the same circular array with the same diameter is measured through the large-view-field lens and the standard lens, the first circle center coordinate and the second circle center coordinate of the same circular array with the same diameter are obtained, composite verification is carried out based on the first circle center coordinate and the second circle center coordinate, distortion verification is carried out on the large-view-field lens, pixel verification is carried out on the standard lens, and the circular array with the same diameter is obtained. And the large-view-field lens and the standard lens are subjected to composite verification, so that the image distortion problem of the double-view-field lens is avoided, and the measurement precision of the double-view-field lens is improved.

The application is suitable for the fields of high-end flexible sensing preparation and multi-modal sensing detection technology, and provides a patch type ultrasonic biomarker integrated sensor and a detection method.The sensor comprises a patch type ultrasonic transducer array, which is used for real-time monitoring of muscle and bone diseases, realizes multi-parameter multi-modal data information fusion detection and analysis of muscle and bone structure-function, a microneedle electrode array, which is used for in-situ detection of biomarkers and collection of physiological electrical signals, a flexible electronic circuit, which is used for bearing and connecting the patch type ultrasonic transducer array and the microneedle electrode array, and an acoustic coupling adhesive layer, which is used for skin fitting and acoustic coupling.The application can systematically evaluate the structure, mechanics, metabolic inflammation level and nerve function activity coordination of muscle and bone, provides a new means for early identification, quantitative evaluation and personalized intervention of muscle and bone diseases, and has important clinical and industrial values.

The present invention is a method and system for quantifying micro cracks inside 3D printed concrete. The quantification method first uses a denoising network and an enhanced attention mechanism denoising network IADNet before performing relevant standard measurements on micro crack images. IADNet can be used to extract features of micro crack images from different fields of view, perceive noise from multiple levels, and perform denoising, which greatly improves image quality, enhances texture details, and is conducive to training segmentation networks. The use of IADNet in conjunction with a semantic segmentation algorithm has the function of finely identifying image information, can quantify micro crack identification, overcome the shortcomings of 3D printed concrete micro crack measurement and analysis, and improve construction efficiency and quality.

The utility model relates to a kind of dual-camera chemiluminescence imaging instrument, including shell assembly, imaging assembly and sample bearing assembly, shell assembly includes shell, shell has cavity;Imaging assembly is located in cavity, imaging assembly includes at least one imaging unit, and imaging unit includes two cameras, and two cameras are black-and-white camera and color camera respectively;Sample bearing assembly is located in cavity, and sample bearing assembly is oppositely arranged with the lens of black-and-white camera and the lens of color camera.This application breaks through the limitation of single imaging mode through the cooperation of black-and-white camera and color camera, improves chemiluminescence detection sensitivity, realizes the efficient collection of multidimensional data, and accurately analyzes complex biological samples.Moreover, black-and-white camera and color camera are integrated in the same cavity, so that the two cameras share the same field of view, avoiding image error caused by sample movement.

The invention discloses a multi-view monitoring video stitching method and system based on energy diagram estimation. The method comprises the following steps: acquiring video data and camera positions of a plurality of monitoring cameras of a road section to be stitched; extracting video frame features of each camera video, dividing pixels in video frames into background pixels or foreground pixels, and matching the background pixels; performing visual depth layering on the video frame by using a multi-layer RANSAC (Random Sample Consensus) method; local matching, layer-by-layer splicing and spliced layer fusion are carried out on video frames which are obtained by two cameras at different positions at the same moment and have overlapped areas layer by layer; and splicing the video frames shot by all the cameras to complete the splicing of the multi-view monitoring video to obtain a panoramic video. The method has the advantages of being small in calculation amount, high in calculation speed and good in splicing effect, the balance of algorithm efficiency and splicing precision is kept, and the method high in efficiency and robustness is provided for the field of surveillance video splicing.

The invention discloses a surface topography measurement method and system based on multi-intensity projection and a medium, and the measurement method comprises the steps: adjusting to form fringe light with different intensities, thereby forming at least four fringe patterns with different phases under each intensity; calculating to obtain original height data of all pixel positions under the current intensity; analyzing the gray value difference of two fringe patterns of which the phase difference is pi at the same pixel position under the current intensity, and carrying out square extraction operation on the sum of squares of a plurality of gray value differences so as to serve as a weight; and performing weighted average to obtain final height data of each pixel position. According to the invention, through a multi-intensity adaptive projection and weighted data fusion mechanism, inherent defects of a traditional phase deflection technology in high dynamic range surface measurement are effectively overcome, full-automatic, high-precision and high-robustness three-dimensional shape measurement is realized, various surfaces with complex reflection characteristics can be intelligently adapted, and the measurement precision is improved. And the practicability, the reliability and the accuracy of the measurement system are remarkably improved.

The present invention relates to the field of visual inspection technology, and discloses a processing method for industrial camera acquisition, comprising the following steps: Step 1, start the industrial camera, initialize basic parameters, enable the intelligent automatic exposure control system, and preliminarily set the exposure time and gain parameters; Step 2, use the developed exposure control algorithm based on image analysis to analyze the initial image data in a low-light environment in real time, and dynamically adjust the exposure parameters of the camera according to the image brightness and details; Step 3, continuously acquire multiple frames of low-light images in a short time. By developing a deep neural network model specifically for low-light environments, it can automatically learn and optimize the features of low-light image processing, effectively improve the clarity and detail performance of the image, and adopt multi-frame information fusion technology to combine multiple frames of low-light image data for image reconstruction and enhancement, which can significantly reduce image noise, enhance image details, and generate clearer and higher-quality image output.

The invention discloses a sugarcane growth monitoring and evaluating device based on multispectral imaging in the technical field of modern farming. The sugarcane growth monitoring and evaluating device comprises an unmanned aerial vehicle, a multispectral acquisition module and a verification image acquisition module, the verification image acquisition module is used for carrying out verification image acquisition on the multispectral image acquisition area, extracting the swaying condition of the sugarcane from the verification image, judging whether the unmanned aerial vehicle needs to stay in the area or not based on the swaying condition of the sugarcane, and when judging that the unmanned aerial vehicle stays, sending the verification image to the unmanned aerial vehicle; and defining an area in which multispectral image acquisition is carried out as a verification area, controlling the multispectral acquisition module to repeatedly carry out multispectral image acquisition on the verification area, extracting the multispectral image of the blocked sugarcane leaf from the verification area, and carrying out position correction on the blocked multispectral image of the leaf. According to the method, the verification image is collected while the multispectral image is collected, the multispectral image is corrected based on the correction image, and the influence of sugarcane leaf shielding on the growth condition judgment and prediction precision is reduced.

Disclosed herein is an organic light emitting diode display device and a driving method thereof, the organic light emitting diode display device comprising: a display panel, in which pixels adjacent to each other are paired and arranged to share a single data line in a pixel area defined by a plurality of gate lines and a plurality of data lines; a gate driver configured to drive the plurality of gate lines; a data driver configured to output a data voltage to a data voltage output channel based on the pixel arrangement; a data switch configured to alternately select pixels and electrically connect the pixels to the data voltage output channel of the data driver; and a timing controller configured to control the data switch as well as the gate driver and the data driver, so that even in a simplified structure of the data driving circuit, the data driving circuit can be driven at a high driving frequency and image quality degradation and image distortion can be prevented.

The application discloses a flying point scanning control method and device, and belongs to the technical field of X-ray scanning imaging. The method is realized based on a flying focal point X-ray source and a static constraint device, and comprises the following steps: acquiring scanning control parameters which are independently set for multiple scanning positions, wherein the scanning control parameters comprise scanning time lengths corresponding to the scanning positions, and different scanning positions can be set as different values; and controlling a focal point switching process of the flying focal point X-ray source according to the scanning control parameters to perform flying point scanning. The application is based on pre-identification of a scanning area, different areas adopt different scanning modes, empty areas can be directly skipped, and key areas are strengthened for scanning. The application reduces a radiation dose, improves scanning efficiency, and improves imaging quality of key areas. However, the existing feedback type scheme cannot distinguish empty areas and metal areas (scattering signals are also weak), and is prone to misjudgment and resource waste.

The invention discloses a head fixing device matched with CT (computed tomography) examination in radiology department, which comprises a bottom plate, a side plate is mounted on the bottom plate, an arc-shaped plate is fixedly connected to the side plate, a first air bag is fixedly connected to the arc-shaped plate, a second air bag is mounted on the bottom plate, and the second air bag is fixedly connected to the arc-shaped plate. The second air bag is communicated with the first air bag through a pipeline; a semicircular limiting plate is fixedly connected to the bottom plate, and scales are arranged on the semicircular limiting plate at equal intervals. The first air bag and the second air bag which are communicated form a linkage supporting structure, air can be automatically distributed according to physiological characteristics of the head and the neck of an examiner, the neck is properly supported, the head is tightly attached to the arc-shaped air bag, and the problems that the neck is suspended, and the height is uncomfortable due to body shape difference of an existing rigid device are effectively solved; the head inclination problem can be quickly found and corrected, and it is ensured that the head is in a preset standard posture of CT scanning.