13 results about "Integral projection" patented technology

The application discloses a particle screenshot method and related equipment, the method comprises the following steps: acquiring a particle image to be processed, and cutting a field of view region from the particle image based on a preset frame selection range; performing integral projection on the field of view region to obtain a one-dimensional sequence group; performing weighted standardization on the initial one-dimensional sequence based on the field of view size of the field of view region in the projection direction to obtain a weighted one-dimensional sequence; performing smoothing processing on the weighted one-dimensional sequence to obtain a smoothed one-dimensional sequence; performing local peak marking based on the smoothed one-dimensional sequence to obtain a local peak position; converting the local peak positions corresponding to two projection directions of the horizontal direction and the vertical direction to obtain a particle position; and cutting a particle screenshot of a target particle from the field of view region based on the particle position. By introducing the weighted standardization link based on the field of view size, and combining the bidirectional integral projection, sequence smoothing and local peak detection, the particle positioning and screenshot can be efficiently and accurately realized. The application can be widely applied in the field of image processing technology.

The invention discloses a particle screenshot method and related equipment, and the method comprises the steps: obtaining a to-be-processed particle image, intercepting a visual field region from the particle image based on a preset frame selection range, carrying out the integral projection of the visual field region, and obtaining a one-dimensional sequence group; performing weighted standardization on the initial one-dimensional sequence based on the view size of the view area in the projection direction to obtain a weighted one-dimensional sequence; smoothing the weighted one-dimensional sequence to obtain a smooth one-dimensional sequence; performing local peak marking based on the smooth one-dimensional sequence to obtain a local peak position; according to the local peak positions corresponding to the transverse and longitudinal projection directions, particle positions are obtained through conversion; and capturing a particle screenshot of the target particle from the view area based on the particle position. The particle positioning and screenshot can be efficiently and accurately realized by introducing a weighting standardization link based on the view size and combining the bidirectional integral projection, the sequence smoothing and the local peak detection. The method can be widely applied to the technical field of image processing.

The application belongs to the technical field of computers and discloses a human eye positioning method and system based on gray scale projection and multi-angle combined HOG features. Firstly, in the image collected by a near-infrared camera, a local adaptive binarization and positioning image centroid combined mode is used to obtain a face region in the image. Then, a gray scale integral projection and least square fitting combined method is used to preliminarily determine all candidate points that may be human eyes. Then, through a gray scale co-occurrence matrix and the maximum connected domain area, most non-human eye candidate points are removed from the preliminarily obtained candidate points, and a gradient image is used to adjust the positions of the candidate points. Finally, through a multi-angle combined HOG feature extraction method, a feature vector of a candidate region centered on the candidate points is extracted, and the feature vector is input into multiple trained classifiers for classification and discrimination. The application has high accuracy and low operation time in the task of positioning human eyes from a face image.

The application provides a kind of drawing feature and digital matching method, system, equipment and medium, comprising: obtaining three-dimensional model and automatically extracting all relevant feature information;Obtain two-dimensional drawing and automatically extract all label information;Overall projection is carried out to three-dimensional model, and the model transformation projection image and projection information corresponding to three-dimensional model are generated;The model transformation projection image and two-dimensional drawing are carried out hash calculation, and the section projection picture position of the three-dimensional model corresponding to two-dimensional drawing is found;According to feature information, label information, two-dimensional drawing and step S3, determine three-dimensional model projection image and projection information, determine which part of feature of three-dimensional model corresponding to label information, and correspond label information to three-dimensional feature list;Output labeled three-dimensional model and feature label data details.The application can improve the efficiency and accuracy of two-dimensional drawing information recognition in production and processing, solve the problem of feature label information that two-dimensional drawing cannot be automatically recognized in automobile production.

The utility model discloses a radar mounting structure, which is characterized in that a radar body and an integrated bulge are spliced together through a splicing part, and the height of the radar is adjusted by operating a lifting mounting structure to adjust the height of a middle mounting block, so that the height of the radar body is adjusted. After installation, the angle of the radar body is adjusted through the angle adjusting button according to actual environment requirements, and the radar body is locked through the locking part after adjustment.

The application discloses a three-dimensional woven preform yarn density detection method and system, and the method comprises the following steps: acquiring a three-dimensional woven preform image; sequentially performing rotation correction, adaptive gamma enhancement and three-dimensional block matching cooperative filtering to obtain a clear image; realizing yarn segmentation through dynamic threshold segmentation and multi-direction morphological optimization; adopting an improved gray integral projection algorithm, generating an adaptive threshold based on global and local gray projection integral mean, binarizing a projection sequence into a square wave signal; extracting yarn quantity and position information by detecting waveform transition points, and calculating yarn density in combination with a space scale coefficient. The method effectively overcomes image tilt, noise and yarn interweaving interference, converts a traditional projection curve with dramatic fluctuations into a stable square wave signal, significantly improves the robustness and accuracy of yarn identification and positioning, and realizes automatic and high-precision detection of high-density three-dimensional woven preform yarn density.

The present application relates to a kind of based on sparse collection's Fourier hologram generation reconstruction method, belong to Fourier hologram generation reconstruction method.Integrated imaging method based on sparse collection is used to generate the stereoscopic meta image of object;According to the stereoscopic meta image of object, multi-view projection incoherent Fourier calculation hologram generation and simulation reconstruction are carried out.The present application has the advantages that compared with the traditional image recognition gray integral projection method, the block matching integral projection method based on SAD search method is used, the parallax error rate obtained by calculation is reduced from 3.125% to 0.2%, the structural similarity SSIM of the generated stereoscopic meta image and the original object view is increased from 0.71 to 0.75, the accuracy of the generated stereoscopic meta image is improved, so that the object Fourier hologram storage information and the reproduced image are more accurate, the block matching integral projection method carries out image correction when running code for block matching, and only searches for matching block in horizontal direction, compared with other algorithms, the steps are more simple.

The invention discloses a cell X-Ray image detection method and device, and belongs to the technical field of digital image processing. The method comprises the following steps: inputting an enhanced X-Ray image into a trained deep learning semantic segmentation model to obtain a binary mask of a bending region; extracting a bending region from the enhanced X-Ray image by using a binary mask of the bending region; carrying out layer separation on the extracted bending region by utilizing the characteristic that the pole piece is in an alternating line segment shape in the enhanced X-Ray image to obtain a multi-layer pole piece image; for each layer of pole piece image, positioning the endpoint coordinates of the pole piece: carrying out gray value integral projection along the extension direction of the pole piece to obtain a one-dimensional projection curve; and searching an inflection point of the projection curve, and positioning an end point coordinate of the tail end of the pole piece based on the inflection point. And the end point of the tail end of each layer of pole piece in the bending area can be positioned.

The invention discloses a three-dimensional woven preform yarn density detection method and system. The method comprises the following steps: acquiring a three-dimensional woven preform image; sequentially performing rotation correction, adaptive gamma enhancement and three-dimensional block matching collaborative filtering to obtain a clear image; yarn segmentation is realized through dynamic threshold segmentation and multi-direction morphological optimization; an improved gray integral projection algorithm is adopted, a self-adaptive threshold value is dynamically generated based on global and local gray projection integral mean values, and a projection sequence is binarized into a square wave signal; the yarn number and position information is extracted by detecting waveform transition points, and the yarn density is calculated in combination with a space proportionality coefficient. According to the method, image inclination and noise and yarn interweaving interference are effectively overcome, a traditional projection curve which fluctuates severely is converted into a stable square wave signal, the robustness and accuracy of yarn recognition and positioning are remarkably improved, and automatic and high-precision detection of the yarn density of the high-density three-dimensional woven preform is achieved.

The invention relates to the technical field of photoelectric measurement and machine vision, and discloses a dynamic rotating speed measurement method based on laser speckles, which is suitable for diffuse reflection characteristic-free rotating parts. Projecting a light beam to a measured object to form speckles, and capturing an asynchronous event stream by using an event camera and a self-adaptive mask; mapping the stream data to a dynamic time window, and converting the stream data into a two-dimensional time surface matrix through a time decay model; resolving a physical rotation center based on known parameters or a low-speed accumulation matrix and feeding back the updated mask; performing logarithmic polar coordinate transformation by taking the center as a pole, and obtaining a roughly estimated angular velocity through integral projection and frequency domain calculation; constructing a search space by taking the displacement as a priori, and executing position compensation and similarity evaluation to extract a refinement rotation displacement; based on this, a second-order state model is constructed, and the instantaneous angular velocity is output through adaptive filtering. According to the invention, the radial vibration interference is decoupled, the extreme working condition range is widened, and the speed measurement precision and the anti-interference capability are remarkably improved.

The invention discloses a digital light processing (DLP) projection lithography system based on a double separation type combined lens and a combined application method thereof, and belongs to the technical field of micro-nano manufacturing. Aiming at the technical bottlenecks of limited function expansion and limited magnification adjustment caused by an integral projection lens in the existing DLP photoetching system, the invention provides a double-separation projection lens structure consisting of an object space front lens group and an image space rear lens group. The front lens group converts reflected light of a digital micromirror array (DMD) into parallel light beams and forms an expansion interface area with an adjustable axial length, and flat plate optical elements such as a beam splitter and a spatial light modulator are allowed to be inserted into the expansion interface area. The rear lens group is integrated on a high-precision micro-displacement platform after being subjected to lightweight design, and continuous fine adjustment of the amplification factor of the system within a certain range can be realized by moving the lens along the axial direction through the micro-displacement platform; or the rolling of the DMD mask pattern and the movement of the whole rear lens group carried by the micro-displacement platform are synchronously coordinated during inclined rolling exposure, so that the pixelate effect is remarkably reduced, and meanwhile, the design complexity of the mask pattern and the requirements of computing resources are effectively reduced.

The invention provides an automatic driving human eye detection method and system based on visual identification, a terminal and a medium, and the method comprises the steps: firstly carrying out the human eye inclination correction of a collected image, and then carrying out the preliminary positioning of the corrected image through an integral projection algorithm. And then, secondary positioning is performed according to the preliminary positioning information by means of the shape model, so that a more accurate human eye detection result can be obtained. According to the method, data are acquired only by depending on a visual sensor, and simplified key data are input into a deep learning network for detection through data processing and screening. According to the method, a complex mapping relation does not need to be constructed, and a large amount of marked data is not needed for network training, so that the operand is remarkably reduced. Moreover, the system detection time is greatly shortened, the detection efficiency is improved, and the harsh requirement for real-time performance in intelligent safe driving research is met.

The invention belongs to the technical field of radar image processing, and particularly relates to an ISAR image target rapid interception method and system, the maximum value and the average value of gray values in integral projection vectors in two directions are calculated, a signal-to-noise coefficient is calculated through the maximum value and the average value of the gray values, and a threshold coefficient is calculated according to the signal-to-noise coefficient; calculating a gray threshold value according to the threshold coefficient and the average value of the gray values; the gray value of each pixel point in the integral projection vector is compared with a gray threshold value, the pixel points which do not meet the gray threshold value are removed, and the pixel points and distance coordinates of the target image are obtained; and carrying out distance-direction gray value summation on the pixel points of the target image to obtain transverse and longitudinal two-end coordinates of the target image. Iterative calculation is avoided, and the processing speed is improved compared with a traditional method. And the anti-noise capability is strong: diffuse noise is suppressed through integral projection, and the target interception integrity is improved.