71results about How to "Achieve objective evaluation" patented technology

The invention discloses an all-reference three-dimensional image quality objective evaluation method based on visual salient feature extraction. According to the method, a left view and a right view of a three-dimensional image pair are processed to obtain a corresponding disparity map; image fusion is performed on the left view and the right view of the three-dimensional image pair to obtain an intermediate reference image and an intermediate distortion image; a spectral residual visual saliency model is utilized to obtain a reference saliency map and a distortion saliency map, and a visual saliency map is obtained through integration; visual information features are extracted from the intermediate reference image and the intermediate distortion image, and depth information features are extracted from the disparity map of the three-dimensional image pair; similarity measurement is performed to obtain measurement indexes of all the visual information features of vision saliency enhancement; and support vector machine training prediction is performed, an objective quality score is obtained, mapping of three-dimensional image quality is realized, and measurement and evaluation of three-dimensional image quality are completed. Through the method, image quality objective evaluation and subjective evaluation have good consistency, and the performance is superior to that of existingthree-dimensional image quality evaluation methods.

The invention discloses an important writing characteristic tagging-based handwritten Chinese character neatness evaluation method. The specific steps include: establishing a Chinese character template library containing importance writing characteristic tagging; recording handwritten Chinese character writing track point set data of a user, and preprocessing the acquired data; obtaining importance tagging writing characteristics of three levels of entire characters, parts and strokes of handwritten Chinese characters of the user; performing matching analysis on the obtained writing characteristics and template characters, so as to obtain a difference relation; and identifying the grade of problems existing in the handwritten Chinese characters of the user according to the obtained difference relation, and performing neatness evaluation on the grade of the problem existing in the handwritten Chinese character of the user according to the grade. The important writing characteristic tagging-based handwritten Chinese character neatness evaluation method provided by the invention can be used for judging whether the handwritten Chinese character writing of the user is neat, and has extremely import significance and value in application of judging handwritten Chinese character writing quality and performing individual guidance of Chinese character writing.

The invention discloses a fusion image quality evaluation method for infrared and visible light images. The method comprises the steps of acquiring an infrared source image, a visible light source image and a fusion image; based on the gradient amplitude information entropy of the fusion image, calculating the global detail characteristic index of the fusion image; dividing the infrared source image into a target region and a background region by adopting the salient region extraction method of the frequency domain, and mapping the division result to the fusion image; respectively calculating the edge structure similarity between the target region of the infrared image and the target region of the fusion image, and the contrast between the target and the background of the fusion image, and calculating the sum of the above obtained results to obtain the local target characteristic index of the fusion image; finally, subjecting the global detail characteristic index of the fusion image and the local target characteristic index of the fusion image to weighted summation so as to obtain the fusion image quality evaluation index of infrared and visible light images. According to the invention, the global detail features and the local target features of the fusion image can be evaluated, while the overall features of the fusion image can also be evaluated. Both the universality requirement and the particularity requirement during the actual evaluation process are taken into consideration. Therefore, the method is high in consistency with the subjective evaluation.

The invention discloses an objective evaluation method of full-reference stereo image quality based on visual salient image feature enhancement. At first, a stereoscopic image is converted from an RGBcolor space to a YIQ color space, The brightness components are extracted from Y-channel to obtain disparity map and visual saliency map, and then the left and right views of Y-channel are fused to obtain the intermediate image, and then the edge / texture and depth information features with significantly enhanced vision are extracted, and similarity measurement is carried out to obtain the corresponding measurement index. Secondly, color information features are extracted from I and Q channels of stereoscopic images, binocular fusion and similarity measurement are carried out, and color information metrics with significantly enhanced vision are obtained. Finally, all the metrics are predicted by support vector regression training, and the objective quality score is obtained. Experiments show that the objective evaluation and the subjective evaluation of the stereoscopic image quality proposed by the invention have good consistency, and the performance is superior to most existing stereoscopic image quality evaluation methods.

The invention provides a petroleum spatial distribution quantitative evaluation chart generation method under reservoir configuration constraint. The method comprises the steps: collecting different types of hydrocarbon source rock samples sequentially according to thermal evolution degree, abundance of organic matter, lithology and organic matter type; carrying out the thermal simulation experiment of the collected samples, quantifying discharged oil and retention oil at each temperature point, employing the organic carbon analysis of residue, Rock-Eval detection, and vitrinite reflectance analysis; drawing an oil discharge rate chart of different types of hydrocarbon source rocks; selecting geological sections of different basins as actual geological samples; densely collecting the hydrocarbon source rocks and storage layer rock core samples and carrying out organic carbon analysis, Rock-Eval detection, vitrinite reflectance analysis and chloroform bitumen quantitative analysis on actual geological sections; drawing an oil discharge rate parameter curve of an actual geological sample; and obtaining the petroleum spatial distribution quantitative evaluation chart generation method under reservoir configuration constraint according to the oil discharge rate charts of different types of hydrocarbon source rocks and the oil discharge rate parameter curve of the actual geological sample.

The invention belongs to the technical field of image processing, and discloses a porous material hole automatic measurement method based on scanning electron microscope image segmentation, which comprises the following steps of 1, initializing parameters; 2, filtering processing; 3, obtaining a potential clustering number through the characteristic value information; 4, calculating histogram information of the filtered image; 5, completing image segmentation by using a fast fuzzy clustering algorithm; 6, extracting data distribution corresponding to the minimum clustering center as hole output to obtain a binary hole image; 7, performing morphological filling operation on the binary hole image; 8, automatically classifying the holes according to areas by utilizing multi-scale morphological closed reconstruction; and 9, counting the number, the area ratio and the actual average area of holes of different grades in the porous material. According to the method, automatic measurement andclassification of the holes of the porous material are realized, the porous material can be objectively evaluated, and an ideal image segmentation effect can be directly obtained.

The invention discloses a wheel set dimension measurement result credibility interval fuzzy evaluation method based on a laser displacement sensor. First of all, a fuzzy integrated evaluation index system is established according to factors affecting wheel set dimension measurement results, and according to requirements for accuracy of the measurement results, wheel set dimension measurement evaluation grades are set; then, through calculating interval properties of membership grades and weight vectors of the measurement result influential factors and analyzing nonlinear features of fuzzy matrix weighting operation, relative dominance of a fuzzy mathematic theory and an interval number vector matrix is introduced, and a wheel set dimension measurement result credibility secondary fuzzy integrated evaluation calculation model is established; and finally, according to the dominance, wheel set dimension measurement result credibility grades are analyzed and determined. According to the method, credibility of a wheel set measurement result can be reasonably evaluated, and the operationality is quite high.

The invention relates to a defense effectiveness evaluation method applied to a network target range. Starting from the severity of the potential attack risk and the two dimensions of two actions of the response of the equipment with the defense function during attack defense, evaluation is realized for design defects of a defense system and problems existing in actual operation of equipment, thedefense effects are quantified, and objective evaluation of defense effectiveness is realized. In application, defense scheme effectiveness evaluation under different application scenes and security risk definition standards can be satisfied, independent modeling analysis does not need to be carried out on each specific scene, the efficiency of the defense strategy of the system can be evaluated in a quantitative form, and compared with a method for simulating a network attack behavior by performing a penetration test by security personnel, the method is more comprehensive and controllable, and is beneficial to clarification of a relationship between the network security equipment and protected assets, a relationship between security threats and defense, and a relationship between the security equipment and an overall defense system.

The invention discloses an image fractal feature based objective image quality evaluation method. The image fractal feature based objective image quality evaluation method includes the following steps: 1, in a preprocessing step, training and fitting an original image in a public knowledge database to acquire basic feature information, namely a fractal dimension matrix and a multi-fractal spectrum, of image quality evaluation; 2, calculating image feature information such as the fractal dimension matrix and the multi-fractal spectrum of a to-be-tested distorted image according to a box counting method; 3, comparing the acquired feature information of the distorted image with the basic reference feature information to acquire differences, and objectively measuring the quality of the distorted image through comprehensive processing according to the differences. The image fractal feature based objective image quality evaluation method has the advantages that high uniformity of objective and subjective evaluation of image quality is achieved, and image quality evaluation performance for specific distortion types of images is excellent.

The invention discloses a Chinese character text line writing quality automatic evaluation method and system based on time sequence point set calculation. The method comprises the specific steps of obtaining point set data of a handwriting Chinese character text line writing track of a user; performing line division according to the obtained point set data; performing word segmentation by adoptingan image processing method according to the line division result to obtain a line axis; calculating user handwritten text line index characteristics by using the line axis description; performing quality evaluation scoring according to indexes in the handwritten text line of the user. The method can be used for judging whether the handwritten Chinese character text line of the user is neat or notand conforms to the general aesthetic principle or not, can be applied to chapter training in hard-tipped pen writing teaching, can also be used for text line personalized writing guidance and the like, and has important practical significance.

The invention relates to an evaluation method for soaping resistance color fastness of yarn-dyed fabrics based on machine vision. According to the evaluation method, overall, elaborate and objective evaluations of soaping resistance color fastness of yarn-dyed fabrics with complex colors and patterns are achieved. The functions of evaluation method for soaping resistance color fastness of yarn-dyed fabrics based on machine vision are achieved by means of a hardware portion and a software portion, wherein the hardware portion can acquire images of the yarn-dyed fabrics, and comprises an industrial video suite (1), a light source (2), a light source controller (3), a computer host (4), a display (5), a data transmission guide line (6), a sample platform (7) for installing the industrial video suite and the light source, and the software portion can achieve acquisition, storage and clipping of images, statistics of grayscale characteristics and grade evaluation of color fastness to washing of the yarn-dyed fabrics and staining of samples of pure cotton fiber white lining are achieved by.

The invention discloses an image definition measuring device and method for a medical display. The image definition measuring device is mainly composed of three modules, namely an image generation module, an optical transfer function tester module, that is an MTF tester module, and an upper computer software module; the upper computer software module is respectively connected with the image generation module and the MTF tester module; and the image generation module mainly comprises an image processing unit, a communication unit and a control unit, wherein the control unit is respectively connected with the image processing unit and the communication unit. The image definition measuring device has the following advantages: the image definition of the pixel level of a medical display panelis quantitatively measured in a numerical value mode through an MTF, so that important reference basis and data support are provided for the objective evaluation and accurate measurement of the display definition of the medical display; the scientificity of the medical display image definition measurement efficiency and the evaluation basis is improved, and unnecessary measurement errors such as human factors are avoided; and the automation degree is high, the deployment is flexible, the maintenance is easy, and the operation is simple.

The invention relates to a soaping-resistance color fastness evaluation method of circular screen printed textiles on the basis of image analysis and realizes the overall, fine and objective evaluation of the soaping-resistance color fastness of the circular screen printed textiles with complicated color pattern structures. The soaping-resistance color fastness evaluation method of the circular screen printed textiles on the basis of machine vision is realized through the cooperation of a hardware part and a software part. The hardware part can realize the image obtaining of the circular screen printed textiles and mainly consists of an industrial video kit (1), a light source (2), a light source controller (3), a computer mainframe (4), a display (5), a data transmission conducting wire (6) and a test sample table (7), wherein the test sample table (7) is used for installing the industrial video kit and the light source. The software part can realize the obtaining, the storage and the cutting of images, the gray-scale feature statistics and the evaluation of the soaping-resistance color fastness grade of the circular screen printed textiles and the staining grade of pure cotton fiber white backing test samples.

The invention discloses an objective non-reference image quality evaluation method based on independent subspace analysis. The method comprises steps that, 1, independent subspace analysis on mass original images in a widely-known database is carried out to obtain a series of relatively independent image characteristics, histogram distribution statistics is carried out, and an edge distribution statistics curve is obtained by employing generalized Gaussian density (GGD) model and is taken as a benchmark reference; 2, image characteristics of to-be-measured distorted images are extracted based on independent subspace analysis, and the GGD model is employed to obtain statistics distribution of the characteristic information; and 3, the distorted image characteristic information statistics distribution obtained through processing and benchmark reference statistics distribution are contrasted, and European style distances corresponding to all integrated characteristic information are measured and accumulated and are taken as quality tolerance of the to-be-measured distorted images. Through the method, objective image quality evaluation and subjective evaluation have good consistency, and the method has better performance compared with traditional image quality evaluation methods.

The invention relates to an image analysis based method for evaluating colorfastness to washing of colored woven cloth, which is used for wholly, finely and objectively evaluating the colorfastness to washing of colored woven cloth with complicated color patterning. The image analysis based method for evaluating colorfastness to washing of the colored woven cloth is realized through the collaboration of a hardware part and a software part, wherein the hardware part can be used for obtaining images of the colored woven cloth, and mainly comprises an industrial video assembly (1), a light source (2), a light source controller (3), a host computer (4), a display (5), a data transmission conducting wire (6) and a sample table (7) used for installing the industrial video assembly and the light source; and the software part can be used for obtaining, storing and cutting the actual images, counting the grayness characteristics, and assessing the grade of colorfastness to washing of the colored woven cloth and the staining grade of a purified cotton fiber white adjacent sample.

The invention discloses a bioreactor virus infected cell quality evaluation method and relates to the technical field of bioreactors, and the method comprises saliency map calculation, saliency map segmentation, connected region marking and infected cell marking. A saliency method is applied to infection quality evaluation for the first time, realizes objective evaluation of virus infected cell states, and solves the problems that errors of subjective evaluation and experience evaluation are large and quantitative analysis cannot be performed.

The invention provides a big-data-based face charming authentication method. The method comprises: at least one storage region is provided at a server to store a photo of a to-be-evaluated object, wherein the photo and the to-be-evaluated object form a one-to-one-correspondence mapping relationship; according to a set rule, two groups of photos are selected, and charming degree evaluation requests are pushed to a plurality of evaluation users; the server receives evaluation results and carries out face farming ranking on to-be-evaluated objects corresponding to the two groups of photos; the above-mentioned steps are repeated and a face charming ranking result of all to-be-evaluated objects is formed; and on the basis of the face charming ranking result, grade classification is carried out by combining a set face charming score curve, so that a face charming score of each face charming grade in the face charming ranking result is obtained. According to the face charming authentication method, on the one hand, data expresses themselves and thus a face charming score and rank of a user is formed naturally, so that manual intervention and analysis can be reduced or avoided; and on the other hand, on the basis of big data accumulation, the face charming ranking result is close to the real information.

The invention relates to the field of colonoscopy process quality evaluation, and provides an enteroscope withdrawal quality intelligent monitoring system and method based on a deep neural network, and the system comprises a quality monitoring model end, a background video processing and calling end, and a video stream display end. The method comprises the following steps: acquiring a video streamgenerated during colonoscopy and splitting the video stream into image frames; detecting a target area in each frame of image and segmenting the target area; judging the effectiveness of the visual field of the segmented target area, and returning the ratio of the number of the images of the effective visual field to the total number of the images to the video stream display end as the visual field efficiency in the lens retreating process; equally dividing the image judged as the effective visual field into four parts according to four quadrants, respectively carrying out lumen, local and fuzzy three-classification judgment, and returning the ratio of the number of images of local classification to the number of images of the effective visual field and the ratio of the number of images of lumen classification to the number of images of the effective visual field to the video stream display end.

The invention discloses an in-vehicle abnormal sound detection and evaluation system, and the system comprises: a plurality of noise signal collection modules which are respectively disposed in a plurality of regions in a vehicle, and are used for the collection of in-vehicle noise signals; a signal noise reduction module which is connected with the noise signal acquisition module and is used for denoising the noise signal to form a denoised signal; an abnormal sound analysis module which is connected with the signal noise reduction module and is used for obtaining various characteristic parameters in the denoised signal; a memory, wherein an abnormal sound feature information base is stored in the memory, and the memory is connected with the abnormal sound analysis module and used for comparing the abnormal sound feature information with the various feature parameters; and a processor, wherein the processor is connected with the memory and is used for determining the abnormal sound evaluation grade of the noise signal. The invention further discloses a method for detecting the abnormal sound in the vehicle, the denoising signal after reconstruction and noise reduction is obtained after the collected noise signal is denoised, the influence of interference factors is eliminated, and the accuracy of abnormal sound detection is improved.

The invention relates to a transfer printing textile soaping color fastness evaluating method based on machine vision, which realizes global, fine and objective evaluation of the soaping color fastness of the transfer printing textile with a complicated color pattern structure. The transfer printing textile soaping color fastness evaluating method based on machine vision is realized by a hardware part and a software part in a coordinating way. The hardware part can be used for obtaining images of the transfer printing textile and mainly comprises an industrial video suite (1), a light source (2), a light source controller (3), a computer mainframe (4), a display (5), a data transmission lead (6) and a test sample bench (7), wherein the test sample bench is used for mounting the industrial video suite and the light source; and the software part can be used for obtaining, saving and cutting images, making a statistic of gray characteristics as well as evaluating transfer printing textile soaping color fastness grades and pure cotton fiber white accompanying test sample staining grades.

The invention relates to an image analysis based method for evaluating colorfastness to washing of digital ink jet printing textiles, which is used for wholly, finely and objectively evaluating the colorfastness to washing of digital ink jet printing textiles with complicated color patterning. The image analysis based method for evaluating colorfastness to washing of the digital ink jet printing textiles is realized through the collaboration of a hardware part and a software part, wherein the hardware part can be used for obtaining images of the digital ink jet printing textiles, and mainly comprises an industrial video assembly (1), a light source (2), a light source controller (3), a host computer (4), a display (5), a data transmission conducting wire (6) and a sample table (7) used for installing the industrial video assembly and the light source; and the software part can be used for obtaining, storing and cutting the actual images, counting the grayness characteristics, and assessing the grade of colorfastness to washing of the digital ink jet printing textiles and the staining grade of a purified cotton fiber white adjacent sample.

The invention discloses an excited response-based no-reference image quality evaluation method. The method comprises the following steps of: 1, constructing an excitation signal, and applying the excitation signal on a to-be-evaluated image to obtain an excitation response signal; converting the to-be-evaluated image and the excitation response signal from an RGB space to a YIQ space; constructinga group of two-dimensional edge detection operators, and carrying out convolution processing on components, on a Y channel of the YIQ space, of the to-be-evaluated image and the excitation response signal, so as to obtain feature information of the components, on the Y channel, an I channel and a Q channel of the YIQ space, of the to-be-evaluated image and the excitation response signal; and taking the feature information as input, and carrying out quality evaluation on image quality mapping of a to-be-tested image by utilizing a machine learning method, so as to obtain an objective quality evaluation value of the to-be-evaluated image. The method is capable of effectively extracting feature information of images, and is high in operation speed and relatively low calculation complexity; on the basis of the method, the objective image quality evaluation values are relatively consistent with subjective evaluation, so that image quality can be well evaluated.

The invention provides a high-precision modeling method and system for attribute data of a layer-control phase-control multi-constraint contaminated site, and the method comprises the steps: buildinga homogenized contaminated site geologic body three-dimensional model through employing a site geologic layered structure and a sedimentary microfacies as constraint conditions; and performing non-uniform segmentation processing on attributes in the attribute range according to the elevation of the contaminated site, and establishing attribute segmentation data; under the constraint of a geologicbody three-dimensional model, establishing a homogenized three-dimensional attribute body model for segmented data, then adooting a spatial interpolation algorithm combining a deterministic model anda non-deterministic model to carry out non-valued area control processing, and establishing a high-precision three-dimensional attribute body model; and finally, performing color separation and colormatching on the model to form a high-precision three-dimensional attribute body model under multiple constraints of layer control and phase control. Through layer control and phase control multi-constraint site pollution attribute modeling, the influence of the concentration distribution diffusivity difference of pollutants between different stratums and different phases on the three-dimensional attribute body model is eliminated, and the pollutant attribute modeling precision is improved.

The invention provides a network service scoring method and device, electronic equipment and a computer readable storage medium. The method comprises the following steps: receiving a network service use request sent by a terminal; searching an adaptive network service from a plurality of pre-deployed network services, and sending data of the network service to the terminal; in the process that theterminal presents the network service according to the data, at least one of watching data representing watching of the network service and behavior data representing operation of the network serviceis acquired; and determining a score of the network service according to at least one of the obtained viewing data and behavior data. Through the method and the device, the network service can be objectively evaluated, and accurate reference information is provided for a user when the user selects the network service.

The invention discloses a flight simulation training evaluation system and method based on multi-source data fusion under research aircraft type, and the system comprises a flight quality monitoring standard model, a standard evaluation monitoring model, and a monitoring project evaluation model, and the flight quality monitoring standard model comprises an FDR / QAR data standard parameter collection module. The standard evaluation monitoring model comprises an under-research aircraft type FDR / QAR data acquisition module, an under-research aircraft type standard matching comparison module and a standard comparison judgment module. According to the invention, the flight quality monitoring standard database is established through the flight quality monitoring standard model, so that the under-research aircraft type standard matching comparison module can conveniently compare the received under-research aircraft type engineering value parameters with the flight quality monitoring standard database and obtain the monitoring project overrun judgment result. And then the monitoring project evaluation model performs monitoring event evaluation according to the project weight value and the event deduction weight value, so that objective evaluation of flight training data is realized, and improvement of flight training and teaching guidance of a pilot is facilitated.

The invention provides a visual and auditory aesthetics evaluation method and system based on electroencephalogram signals, and relates to the technical field of brain cognitive neural functions. The visual and auditory aesthetics evaluation method comprises the following steps: collecting original electroencephalogram signals of a subject during aesthetic activities, and obtaining EEG original data of the subject; preprocessing the obtained EEG original data to obtain pure EEG signals with the same duration under different conditions; calculating energy density spectrums of five different frequency bands of delta, theta, alpha, beta and gamma through short-time Fourier transform, and obtaining energy density spectrum feature matrixes of the different frequency bands; and screening an optimal feature by using a recursive feature based on a support vector machine to perform classifier training, performing classification identification on aesthetic subjective judgment of the participant, obtaining a subjective judgment result of the participant, and feeding back the subjective judgment result to the participant through sound and picture reminding. According to the method, high-precision identification of aesthetic judgment of participants can be realized, objective evaluation of visual and auditory aesthetic can be realized, and thoughts are developed and a technical foundation is laid for fairness of aesthetic evaluation.

The invention provides a respiratory effect measurement method and system, and relates to the technical field of image quality testing, and the method comprises the steps: collecting a video, carryingout the scene classification of the video according to the image complexity, and obtaining a plurality of test scenes; obtaining an image of each test scene, and calculating definition information ofthe image by adopting a definition algorithm; evaluating the respiratory effect of the test scene according to the definition information to obtain a respiratory effect evaluation value; and carryingout weighted calculation on the respiratory effect evaluation values corresponding to the plurality of test scenes according to a preset weight to obtain a respiratory effect comprehensive evaluationvalue. According to the method, objective evaluation of the respiratory effect can be realized, and the accuracy, repeatability and economy of the respiratory effect test are improved.

The invention discloses a credit evaluation method and system, which includes obtaining the initial confidence value of each user according to the user's personalized credit control information; analyzing the user's detailed list data to obtain the user's social circle information; using the obtained social circle The circle information calculates the confidence value correction factor, and corrects the initial confidence value to obtain the user's credit value. Through the method of the present invention, the credit parameters of the user's social circle are introduced into the user's credit evaluation system, and the big data platform is used to mine and calculate the information of the social circle, and the confidence correction factor of the social circle in which the user is located is introduced on the basis of the user's own initial confidence value , which realizes the objective evaluation of the user's credit, which is beneficial to the operator to comprehensively evaluate the user's credit reliability from multiple dimensions, thereby improving the user experience, reducing the operator's business risk, that is, reducing the probability of events such as arrears.

The invention discloses a customer service quality evaluation method, device and equipment based on natural language processing. The method comprises the following steps that: obtaining communicationinformation between a network customer service and a user, adopting a natural language processing mode to extract keyword from the communication information, normalizing the keyword, forming feature data of the keyword, carrying out similarity comparison on the formed feature data and preset feature information representing the customer service quality of the network customer service to obtain a similarity comparison result, and evaluating the customer service quality of the network customer service according to the similarity comparison result. By means of the mode, the quality of customer service provided by the network customer service can be objectively evaluated.

The invention relates to a photovoltaic prediction precision assessment method and device giving consideration to the sun position, and the method comprises the steps: defining an installed capacity proportion threshold value p at an evaluation-free moment; calculating the maximum theoretical output of photovoltaic power generation; judging a relative error of prediction through the product of the maximum theoretical output of photovoltaic power generation and the installed capacity proportion threshold value p; obtaining prediction precision evaluation index of photovoltaic power; and completing the evaluation of the photovoltaic prediction precision evaluation index through comparison. The method can effectively reflect the sizes of photovoltaic power generation prediction errors in different time segments, can grade the quality of different prediction results, and obtains the objective prediction precision evaluation result. The method effectively improves the prediction precision of photovoltaic power.