33 results about "Standard test image" patented technology

Masks that include optical proximity correction or phase shifting regions are increasingly being used in the manufacturing process. These masks, either initially or after repair, can have “soft” defects, e.g. phase and/or transmission defects. In accordance with one feature of the invention, soft defect information can be computed from standard test images of a mask. This soft defect information can be used to generate an accurate simulated wafer image, thereby providing valuable defect impact information to a user. Knowing the impact of the soft defect can enable a user to make better decisions regarding the mask. Specifically, a user can now with confidence accept the mask for the desired lithographic process, repair the mask at certain critical locations, or reject the mask, all without exposing a wafer.

A focus-quality judging method includes the following steps. Firstly, a comparing object including a target area is provided. A reference image pickup device is provided to shoot the comparing object to obtain a standard image including a standard target area image. Then, a first pixel number contained in the standard target area image is counted. The comparing object is shot by a test image pickup device to obtain a test image including a test target area image, wherein the test image has the same resolution as the standard image. Then, a second pixel number contained in the test target area image is counted. The first pixel number contained in the standard target area image is compared with the second pixel number contained in the test target area image. According to the comparing results, it is discriminated whether the test image pickup device performs an accurate focusing operation.

The invention discloses a single-image super-resolution reconstruction method used for solving the technical problem that image reconstruction precision is low in an existing image super-resolution reconstruction method. According to the technical scheme, firstly, high/low-resolution image blocks are extracted through a great number of high-resolution images and serve as a dictionary; then, according to input image blocks, the low-resolution image blocks in the dictionary are selected to conduct calculation of a deformation field; finally, the corresponding high-resolution image blocks in the dictionary are deformed. A final high-resolution image is obtained through local restriction and global restriction. By means of the deformable image blocks, the expression capability of the dictionary is greatly enhanced, and therefore the final reconstruction effect is improved. The 30000 7*7 image blocks are selected to serve as the dictionary. When the extraction step length of the image blocks satisfies the equation that S=1, and super-resolution reconstruction with the enlargement factor being 3 is conducted on a 256*256 standard test image which is a Lena image, reconstruction precision satisfying the equation that PSNR=31.53 can be achieved and is higher than the reconstruction precision satisfying the equation that PSNR=29.68 in documents.

The embodiment of the invention discloses an image recognition method, a terminal device and a computer readable storage medium. The method comprises the following steps: determining image interference information, and removing the interference information in an input source image based on the image interference information to obtain a standard image; and determining a plurality of groups of imageprocessing parameters corresponding to the image interference information, and processing the standard image according to each group of image processing parameters to obtain a plurality of feature images to be processed; and performing feature extraction on the plurality of to-be-processed feature images through a multi-channel feature extraction model to obtain a plurality of image features, anddetermining a target image feature from the plurality of image features; and inputting the target image features into an image classifier, and determining a target object in the input source image through the image classifier. By adopting the embodiment of the invention, the accuracy of the image recognition method can be improved, the complexity of the image recognition method is reduced, and the applicability of the image recognition method is improved.

The invention relates to the technical field of pattern recognition, in particular to a method for generating sample images under all vision angles on the basis of a standard image. The method comprises the following steps: firstly, a plurality of parameters are adopted to determine the space three-dimensional positions where a camera is located when in shooting, all values of the parameters are combined to be capable of covering all the positions where the camera can be located in the three-dimensional space, and each group of values only corresponds to a certain position in the space; then two groups of different parameters corresponding to two different space positions of the camera are used for showing the transformation relation among real images shot by the camera at the two positions; and then the transformation relation is applied to the real images shot at the first camera position, and a sample image can be generated. The image transformation relation leads the generated sample image to be capable of simulating the real images shot from corresponding vision angles completely, and can generate needed image under specified vision angle by controlling the values of the parameters.

The invention discloses an acquisition method of a standard image in automated optical inspection, and automated optical inspection equipment. The method comprises the following steps: continuously acquiring a scanned image of a to-be-inspected object; acquiring the difference between the scanned image and the current standard image; determining whether to update the current standard image according to the difference; and updating the current standard image by using the scanned image if needing to update the current standard image, thereby obtaining the final target standard image. Through themethod disclosed by the invention, the situation that the obtained target standard image is free from defect can be guaranteed, and then a detection effect of performing defect detection by using thetarget standard image can be improved.

The invention discloses a texture image identification method which includes the following steps: preprocessing a texture image and extracting a character area; performing coordinate-system conversion on the character area and calculating a projection histogram; performing Fourier transformation on the projection histogram so as to obtain the characteristic vector of the texture image; and executing the above-mentioned steps on a to-be-identified texture image and a standard texture image so that the characteristic vector of the to-be-identified texture image and the characteristic vector of the standard texture image are obtained and calculating the similarity between the characteristic vector of the to-be-identified texture image and the characteristic vector of the standard texture image and judging whether the to-be-identified texture image is a standard texture image according to a similarity threshold. The invention also discloses a texture image identification device. The texture image identification method and identification device are great in identification effect of dynamic objects, high in interference resistance, low in identification missing rate, particularly low in error identification rate, simple in calculation, great in instantaneity and capable of satisfying real-time target detecting tasks under a dynamic scene.

The invention discloses a projection equipment optical performance measurement device and method based on motion control. The device comprises a control processing system, a three-dimensional motion mechanism and a chroma meter, wherein the chroma meter is mounted on the three-dimensional motion mechanism, the control processing system is used for sending standard test image data to projection equipment, sending a motion control instruction to the three-dimensional motion mechanism, acquiring measurement data of the chroma meter and processing the measurement data to acquire measurement data in consistent with comparison conditions, the three-dimensional motion mechanism is used for doing motion according to the motion control instruction and making the chroma meter move to a designated point to carry out optical performance measurement on the projection equipment, and the chroma meter is used for carry out optical performance measurement on the projection equipment to acquire measurement data. The measurement device and the measurement method are advantaged in that optical performance of the projection equipment can be automatically measured, and optical performance measurement cost of the projection equipment is reduced.

The invention discloses a real-time image correction method based on a cell phone platform. The method comprises the following steps of: 1, capturing a standard test image and calculating a correction matrix by using the standard test image; 2, storing the correction matrix in an online database for real-time downloading; 3, inputting the height and the arm length of a cell phone user to obtain a cell phone photographing height, downloading all the corresponding correction matrix data from the online database according to the photographing height, and storing all the corresponding correction matrix data in the local memory of the cell phone; 4, subjecting the image captured by the cell phone to coordinate transformation: searching a correction matrix corresponding to the angle information acquired by a direction sensor in real time in the local memory of the cell phone, and transforming the coordinate of each pixel in the image captured by the cell phone in real time into the coordinate of each pixel in a corrected image according to the correction matrix and a given transformation relation; and 5, performing grayscale interpolation calculation to obtain the grayscale values of respective pixels in the corrected image with transformed coordinates.

The invention discloses an image test method and an image test device. The method comprises the following steps: acquiring a test image and a standard image corresponding to the test image; extracting the feature points of the test image and the feature points of the standard image through a scale-invariant feature transform algorithm, and determining a first number of matching feature points from the feature points of the test image and the feature points of the standard image; removing mismatching points from the first number of matching feature points through a random sampling consensus algorithm to get a second number of matching feature points, judging whether the test image and the standard image match each other according to the first number and the second number, and determining a homographic matrix between the test image and the standard image according to the second number of matching feature points after matching; and determining the tensile state of the test image at the time of display according to the homographic matrix.

The invention discloses a machine vision label proofreading method based on vector object segmentation. The method comprises the following steps: 1, obtaining a standard sample grid image group by performing analyzing and storage processing on a standard sample image of a label, wherein the standard sample image comprises a standard sample vector figure, a standard sample vector text and a standard sample grid mage; 2, obtaining a sample image to be detected of a label to be detected for comparing with the standard sample grid image group so as to obtain difference images different from the standard sample image in the sample image to be detected; and 3, by taking a storage unit as a unit, presenting a proofreading result to an operator so as to complete proofreading. The method automatically processes vector figures and vector texts, prevents occurrence of the difference images not needing attention in the proofreading, and enables the proofreading result to present real differences between a sample to be detected and a standard sample.

The invention discloses a display screen visual angle measurement method based on image recognizability, and relates to the technical field of visual angle measurement. The method comprises the following steps: 1: carrying out equipment installation and test preparation; 2, inputting a standard test image to the tested display screen through a signal generator; 3, changing the measuring point leftwards or rightwards by taking the point P0 as a reference measuring point; 4, carrying out identification degree measurement; 5: carrying out four component function synthesis on the identifiable degree VDL; and 6, carrying out value taking based on the steps, and completing measurement of the visual angle of the display screen. According to the display screen visual angle measurement method based on the image recognizability, the display screen visual angle measurement operation is accurate and convenient, the adaptability is high, the image resolution is high, and the image quality guarantee of display screen watching is obtained by calculating the optimal visual angle.

The invention provides a visual fusion intersection angle measuring method and device and a storage medium, which can enable the eyes of a user to be in a relaxed state when the user watches a 3D video or carries out visual training through the 3D video, and improve the user experience. The method comprises the following steps of: displaying a standard test 3D image through a display device, wherein the standard test 3D image is a 3D test image formed by fusing a first image and a second image; offsetting at least one of the first image and the second image according to a user operation instruction; if a recording instruction of the user is received, recording a target offset between the first image and the second image according to the recording instruction; determining the pupil distance of the user and the vertical distance between the pupil and the display device; and calculating the visual fusion intersection angle of the user according to the target offset, the pupil distance of the user and the vertical distance.

The invention relates to a human medical image organ geometric comparison detection method. The method comprises the following steps: non-colinear multiple matched feature points are determined on a to-be-detected anatomic image and a standard image in a module matching mode; based on the feature points, a three-dimensional coordinate system for the standard image and a three-dimensional coordinate system for the to-be-detected anatomic image are built respectively; in the three-dimensional coordinate system, multi-resolution analysis is carried out on the standard image and the to-be-detected anatomic image respectively, and sub images with multiple stages of resolution are obtained; and registration is started from a sub image with the lowest resolution, registration is carried out stage by stage until registration parameters for the image under the highest resolution are obtained. The method of the invention can be used for human medical image organ geometric comparison.

The invention provides a display correction parameter extraction device. The display correction parameter extraction device comprises a communication interface module used for reading actual measurement data; a standard data generation module used for generating standard working data; a standard test image generation module used for generating a standard test image; an adjusting mode processing module used for selecting a to-be-corrected component to be corrected and sending actual measurement data and standard working data related to the to-be-corrected component to the PID adjusting module;a PID adjustment module used for carrying out PID difference operation on the actual measurement data and the standard working data related to the component to be corrected, and carrying out increment; and the adjustment mode processing module is also used for summing the increment and the to-be-corrected component to obtain a corrected test image and generating a correction parameter lookup table. The PID controller has the advantages of mutually independent control parameters, simple parameter selection and the like, and a mathematical model of a controlled object does not need to be provided, so that the adaptability is high, the operation is simple, and the production benefit is improved.