447 results about "Co-occurrence matrix" patented technology

A finger sensing device includes a finger sensing area, and a processor cooperating therewith for reducing a number of possible match combinations between a sensed finger data set and each of a plurality of enrolled finger data sets. The processor may reduce the number of possible match combinations by generating a plurality of overlap hypotheses for each possible match combination, generating a co-occurrence matrix score based upon the plurality of overlap hypotheses for each possible match combination, and comparing the co-occurrence matrix scores to thereby reduce the number of possible match combinations. The processor may also perform a match operation for the sensed finger data set based upon the reduced number of possible match combinations. The sensed finger data set may include a sensed finger ridge flow data set, and each enrolled finger data set may include an enrolled finger ridge flow data set.

The invention discloses a crowd density estimation method based on video analysis and a pedestrian volume statistical method based on the video analysis. The crowd density estimation method includes the flowing steps of (1) off-line training: manually counting crowd density data, extracting characteristics and conducting training; and (2) on-line estimating: extracting the characteristics and conducting regression prediction by utilizing trained model parameters. The pedestrian volume statistical method includes the step of setting up a robust relationship between a scene and a line-passing number of people by combing the crowd density and a micro-region pedestrian flow speed before a line is passed. Characteristics such as foregrounds, edges and gray scale co-occurrence matrixes are extracted based on a whole area to conduct crowd density estimation, problems of dense crowds, sheltering and the like can be well solved through mixing of the characteristics, and real-time crowd density estimation is achieved. In addition, on the basis of area crowd density estimation, pedestrian volume estimation is conducted through combination of the pedestrian flow speed based on an optical flow, detection and tracking of a large number of individuals under a complex environment are avoided, and two-way pedestrian volume counting of accurate robust under dense crowds is achieved.

The invention discloses a strip steel surface area type defect identification and classification method which comprises the following steps: extracting strip steel surface pictures in a training sample database, removing useless backgrounds and keeping the category of the pictures to a corresponding label matrix; carrying out bilinear interpolation algorithm zooming on the pictures; carrying out color space normalization on images of the zoomed pictures by adopting a Gamma correction method; carrying out direction gradient histogram feature extraction on the corrected pictures; carrying out textural feature extraction on the corrected pictures by using a gray-level co-occurrence matrix; combining direction gradient histogram features and textural features to form a feature set, which comprises two main kinds of features, as a training database; training the feature data with an improved random forest classification algorithm; carrying out bilinear interpolation algorithm zooming, Gamma correction, direction gradient histogram feature extraction and textural feature extraction on the strip steel defect pictures to be identified in sequence; and then, inputting the feature data into an improved random forest classifier to finish identification.

Provided is a haze monitoring method based on computer vision. The monitoring method includes the steps: collecting data of pre-set far and near target regions which are dark in color in a scene, and giving haze monitoring results based on comparison of the computation of target object visual features and sample images under different haze conditions. Visual features expressing an image comprises color features containing pixel color saturation mean value and blue component mean value, shape features containing feature point number and edge pixel number, textural features containing grey level co-occurrence matrix features and wavelet transform sub-band coefficients, and feature vectors expressing differences between a far target object and a near target object. According to the haze monitoring method based on computer vision, a direct measurement method by means of visual features which is closely associated to manual observation of the haze in principle is put forward, comprehensive monitoring of haze conditions in an entire region is achieved easily, and high-precision monitoring results can be guaranteed with enough sample data.

Method for detecting textural defects in an image. The image, which may have an irregular visual texture, may be received. The image may be decomposed into a plurality of subbands. The image may be portioned into a plurality of partitions. A plurality of grey-level co-occurrence matrices (GLCMs) may be determined for each partition. A plurality of second-order statistical attributes may be extracted for each GLCM. A feature vector may be constructed for each partition, where the feature vector includes the second order statistical attributes for each GLCM for the partition. Each partition may be classified based on the feature vector for the respective partition. Classification of the partitions may utilize a one-class support vector machine, and may determine if a defect is present in the image.

The invention relates to a texture-based insulator fault diagnostic method. According to the invention, a visible light image collected in the inspection process of a high voltage transmission line by a helicopter is used as an object to be processed, and the diagnosis can be carried out based on an insulator fault of the visible light image. The method comprises the following steps of: inputting an insulator image, carrying out gray processing, obtaining a bounding rectangle and rotating, carrying out a GLCM (gray level co occurrence matrix) method, blocking, obtaining textural features, carrying out Gabor filtering, blocking, calculating block-mean value and variance, performing feature fusion, and determining whether to have a string-drop phenomenon based on a threshold value. The method provided by the invention diagnoses the insulator string-drop characteristic by texture, integrates the thoughts of the most classical GLCM texture diagnostic method in the texture diagnosis and the recent research focus Gabor filter texture diagnosis, adjusts the parameter settings of the GLCM and the Gabor filter and efficiently and accurately finds out the string-drop insulators. The method can effectively improve the efficiency of the thermal defect detection of the power transmission line and can be effectively applied to the inspection business of the vehicle-mounted or helicopter power transmission line.

The invention provides a weld surface defect identification method based on image texture. Miniature CCD camera photographing parameters are set according to the acquisition image standard; an acquired true color image is converted into a grey-scale map, and a gray scale co-occurrence matrix is created; 24 characteristic parameters in total, including energy, contrast, correlation, homogeneity, entropy and variance, are respectively extracted in the direction of 0 degree, 45 degrees, 90 degrees and 135 degrees, and normalization processing is performed on the extracted characteristic parameters; a BP neural network is trained by utilizing a training sample image, and the number of neurons of the neural network, a hidden layer transfer function, an output layer transfer function and a training algorithm transfer function are set; test sample characteristic parameters are outputted to the trained BP neural network to perform classification and identification; and the matching degree of the test sample and different types of surface welding quality training samples is calculated so that automatic classification and identification of the test sample surface welding quality can be completed.

The invention discloses a band steel surface defect feature extraction and classification method, and belongs to the fields of mode recognition and image processing. The band steel surface defect feature extraction and classification method comprises the steps: extracting a reference sampling size chart of a band steel surface defect sample database; obtaining a reference sampling image, and constructing a gradient size and direction co-occurrence matrix; by aiming at a defect inner area of the reference sampling image, constructing a grayscale size and direction co-occurrence matrix; generating a feature vector sample training library; trimming a training sample set and extracting a multiplying factor by a method of combining K-nearest neighbour with R-nearest neighbour; improving a classifier by using a multiplying factor of the trimmed sample; obtaining a multi-class classifier model; according to the reference sampling size chart, converting the defect test sample into a reference sampling image, then extracting a 25-dimensional feature quantity, inputting the 25-dimensional feature quantity into the multi-class classifier model, and finishing the defect automatic recognition. According to the band steel surface defect feature extraction and classification method, the scale and rotation are not changed, the influence by other adverse factors is restrained, and recognition efficiency and accuracy are improved.

The present invention discloses an image recognition-based facial skin analysis system. The image recognition-based facial analysis system is capable of conducting a series of evaluation on the condition of the skin to obtain an overall evaluation. The system mainly comprises a face detection and segmentation module, a whitening degree detection module, a roughness detection module, a stain amount detection module and a comprehensive judgment module. Firstly, a detection image is input to be subjected to front face detection, area location and segmentation by a classifier. Secondly, the segmented image is subjected to skin color brightness and ruddy condition analysis, wherein the roughness of a feature index quantification detection area is obtained based on a gray level co-occurrence matrix. Thirdly, the stain amount is analyzed based on the similar watershed method through calculating the sum of sectional areas. Finally, the comprehensive judgment is conducted according to all detection results. Based on the technical means, the application effect of a product is quantified, so that the quality of the product is controlled. The system has great economic value for cosmetic manufacturers. Meanwhile, the system brings more economic benefits.

The invention discloses a method and a system for identifying an offline handwritten signature. The method comprises the following steps of S1, preprocessing an offline signature sample in an offline signature sample library; S2, extracting a plurality of characteristics for a preprocessed signature image, wherein the characteristics include moment characteristics, local binary pattern characteristics, gray-level co-occurrence matrix characteristics and pulse coupled neural network characteristics; S3, training the plurality of extracted characteristics of the offline signature sample to obtain a trained standard sample library; S4, acquiring a signature to be detected, and preprocessing the signature to be detected to obtain the plurality of characteristics of the signature to be detected; and S5, matching the plurality of characteristics of the signature to be detected with corresponding characteristics of the offline signature sample in the standard sample library, and identifying whether the signature to be detected is a genuine signature or a forged signature. The method and the system can effectively and accurately identify an offline signature to be detected.

The invention discloses a CT image liver segmentation method and system based on characteristic learning, being able to effectively improve the segmentation precision of the liver in a CT image. The method comprises the steps: S101: reading a training image set and an image to be segmented, wherein the training images in the training image set and the image to be segmented are the CT images of a belly; S102: extracting the Haar characteristic of the training images and the image to be segmented, the local binary pattern characteristic, the directional gradient histogram and the co-occurrence matrix characteristic; S103: utilizing a principal component analysis method to perform characteristic fusion on all the extracted characteristics so as to acquire more effective characteristics; S104: utilizing a classifier to classify the characteristics of each pixel of the image to be segmented to obtain a liver probability graph; and S105: combining the liver probability graph with the image to be segmented, modifying the graph model weight of a random walk segmentation algorithm to realize segmentation of the liver.

A system and method for training a statistical machine translation model and decoding or translating using the same is disclosed. A source word versus target word co-occurrence matrix is created to define word pairs. Dimensionality of the matrix may be reduced. Word pairs are mapped as vectors into continuous space where the word pairs are vectors of continuous real numbers and not discrete entities in the continuous space. A machine translation parametric model is trained using an acoustic model training method based on word pair vectors in the continuous space.

The invention provides a dense population estimation method and a system based on multi-feature fusion. The method comprises the following steps: partitioning an image into N equal sub-blocks; performing hierarchical background modeling on the image by using a method based on a CSLBP (Center-Symmetric Local Binary Pattern) histogram texture model and mixture Gaussian background modeling, extracting the foreground area of each sub-block subjected to perspective correction, detecting the edge density of each sub-block in combination with an improved Sobel edge detection operator, and extracting four important texture feature vectors in different directions for describing image texture features in combination with CSLBP transform and a gray-level co-occurrence matrix; performing dimension reduction processing on the extracted population foreground partition feature vectors and texture feature vectors through main component analysis; inputting the dimension-reduced feature vectors into an input layer of a nerve network model, and acquiring the population estimation of each sub-block through an output layer; adding to obtain the total population. The dense population estimation method and system have high accuracy and high robustness, and a good effect is achieved in the population counting experiment of subway station monitoring videos.

The invention relates to a monitoring method based on image features and an LLTSA algorithm for a tool wear state. According to the method, an image texture feature extraction technology is introduced into the field of tool wear fault diagnosis, and monitoring for the tool wear state is realized in combination with three flows of ' signal denoising', 'feature extraction and optimization' and 'mode recognition'. The method comprises the steps of firstly, acquiring an acoustic emission signal in a tool cutting process through an acoustic emission sensor, and carrying out signal denoising processing through an EEMD diagnosis; secondly, carrying out time-frequency analysis on a denoising signal through S transformation, converting a time-frequency image to a contour gray-level map, extracting image texture features through a gray-level co-occurrence matrix diagnosis, and then further carrying out dimensionality reduction and optimization on an extracted feature vector through a scatter matrix and the LLTSA algorithm to obtain a fusion feature vector; and finally training a discrete hidden Markov model of the tool wear state through the fusion feature vector, and establishing a classifier, thereby realizing automatic monitoring and recognition for the tool wear state.

The invention provides a space domain and frequency domain characteristic based parallel SAR (synthetic aperture radar) image classification method. The method includes: by combining space domain and frequency domain characteristics of SAR images and based on the parallel computation environment, dividing the SAR images into n blocks prior to selecting a small image block in the size of 8*8 pixels around each pixel from each image block, computing corresponding wavelet energy features, gray-level co-occurrence matrix features and filtered gray-level average features of each pixel in each small image block, recovering the wavelet energy features, gray-level co-occurrence matrix features and filtered gray-level average features of each pixel in the n small image blocks to obtain wavelet energy features, gray-level co-occurrence matrix features and filtered gray-level average features of each pixel in the SAR images, forming the features into feature vectors for clustering, and finally classifying the SAR images. According to the method, quick classification of the SAR images depends on efficient information processing capability of a parallel cluster computer system, quick classification is realized, and the problem of low speed of SAR image classification in large data volume is solved.

The invention discloses a word vector model based on point mutual information and a text classification method based on CNN. The method comprises the following steps: (1) training a word vector modelthrough a global word vector method based on point mutual information; (2) determining a word vector matrix of the text according to the trained word vector model; (3) extracting features from word vector matrix by CNN and training classification model; (4) extracting input text features according to the trained word vector model and CNN feature extraction model; (5) according to the text featuresextracted from CNN feature extraction model, calculating the mapping distance between text and preset categories by softmax and the cross entropy method, wherein the nearest one is the correspondingcategory of text. This method overcomes the shortcomings of Glove word vector in the semantic capture and statistical co-occurrence matrix, reduces the training complexity of the model, can accuratelymine the text classification features, is suitable for text classification in various fields, and has great practical value.

An image recognition method of coal and gangue is disclosed. According to differences of gray scales and textures of coal and gangue surfaces, through using a computer image processing technology, 5 characteristic parameters of a gray histogram of coal and gangue images and 11 characteristic parameters of a gray level co-occurrence matrix are extracted respectively, totaling 16 characteristic parameters. A BP neural network algorithm is used to identify the coal and the gangue and process the coal and gangue images. The computer is used to carry out intelligent identification and sorting. In a trained BP neural network, any one of the characteristic parameters extracted from the coal or gangue digit images is input so that an identification result can be obtained from an output layer. The method is beneficial to sort the coal and the gangue. Coal quality can be increased. A recognition rate is high and operation is easy.

The invention discloses an illegal parking detection method based on image texture. The method sequentially includes the following steps of obtaining the background, segmenting targets and analyzing the texture. The background obtaining is mainly achieved through the regional division of a determined detection area and a detection area, the establishment of a detection area model and the background renewing. The target segmenting is mainly achieved through the dynamic threshold value selection, the target segmentation and the morphology filtering. By means of the target segmentation method where dynamic threshold values are adopted, moving targets can be more accurately obtained through segmentation; meanwhile, the noise interference generated in the segmentation process can be effectively eliminated through the opening operation in the morphology filtering; the texture analyzing is mainly achieved through the static target detection, the running and parked vehicle recognition and the illegal parking event conforming; whether static objects appear or not is judged by analyzing the process during which the image gray value reaches the steady state; meanwhile, the false alarm existing in the process is eliminated through the entropy of an image gray level co-occurrence matrix.

The invention discloses an undesirable image detecting method based on connotative theme analysis, which is substantially used for solving the problem of wrong judgment on normal images resulting from semantic information consideration failure in the present undesirable information detecting method. The scheme is as follows: extracting a skin region of an image by a double-blending Gaussian model; generating a codebook base containing distinguishing features in the skin region by a word bag model, and representing each training image to a group of word co-occurrence vectors with weights via aword frequency-inverse identification file frequency method; forming all co-occurrence vectors to a co-occurrence matrix, performing LDA model creation on the co-occurrence matrix to obtain the themeof the image; inputting the mixed theme of the training image in a BP neural network to train an undesirable image classifier; and obtaining the theme of an image to be measured, inputting the theme to the undesirable image classifier, and judging whether the theme is an undesirable image so as to finish the undesirable image detection. As shown in the test, the invention can be used for better distinguish the undesirable images and the normal images, so that the invention can be used for filtering the erotic information in the images.

In various embodiments, the present invention relates disparate objects based on user behavior, thus enabling search engines to provide more comprehensive and accurate results. According to various embodiments of the present invention, multiple kinds of interactions by users with multiple classes of objects can be analyzed. The result is that disparate classes of objects can be related. Derived relations between text and objects can be used to implement search-like functionality or to extend a conventional text retrieval system.In one embodiment, the present invention is used to improve search results and / or recommendations by employing a filtered co-occurrence matrix that provides a representation as to which queries tend to co-occur with the originally submitted query. By supplementing or replacing the original query with co-occurring queries, the system of the present invention is able to generate results that are more likely to be of interest.

A method for the automated analysis of digital images, particularly for the purpose of assessing the presence and severity of cancer in breast tissue based on the relative proportions of tubule formations and epithelial cells identified in digital images of histological slides. The method includes the step of generating a property co-occurrence matrix (PCM) from some or all of the pixels in the image, using the properties of local mean and local standard deviation of intensity in neighbourhoods of the selected pixels, and segmenting the image by labelling the selected pixels as belonging to specified classes based upon analysis of the PCM. In this way relatively dark and substantially textured regions representing epithelial cells in the image can be distinguished from lighter and more uniform background regions Other steps include identifying groups of pixels representing duct cells in the image based on intensity, shape and size criteria, dilating those pixels into surrounding groups labelled as epithelial cells by a dimension to correspond to an overall tubule formation, and calculating a metric based on the ratio of the number of duct pixels after such dilation to the total number of duct and epithelial pixels. Other uses for the method could include the analysis of mineral samples containing certain types of crystal formations.

The invention discloses a remote sensing image processing method combined with the shape self-adaption neighborhood and the texture feature extraction for image preprocessing. The method includes subjecting compressed image to a gray level co-occurrence matrix calculation; subjecting the generated gray level co-occurrence matrix to S coefficient modification of an SAN (Storage Area Networking) irregular object window to obtain a regular matrix; calculating a new co-occurrence matrix according to the modified regular matrix and selecting texture descriptors with obvious feature and low correlation; extracting texture feature map in the SAN irregular images; and calculating to obtain accurate images with combination feature which is overall comprehensive feature of neighborhood. According to the method, the overall classification accuracy based on a shape self-adaption neighborhood method can be improved by 4%. The method can not only extract the texture feature in the SAN irregular images of remote sensing images completely, but also process the extraction of mixed pixel feature of the fuzzy edge of earth surface objects, and is applicable to texture extraction of earth surface objects in natural states.

The invention discloses a holographic touch interactive exhibition system with multisource input and intelligent information optimizing functions. The system senses surrounding environment information by the aid of a plurality of sensors, a smoke sensor is started to monitor fire without audiences, faces in videos are recognized when the audiences are present, proper broadcasting information is recommended, and the identity sense of customers is enhanced. The exhibition system introduces an interactive mode, and the audiences can realize man-machine interaction by the aid of a holographic touch screen to acquire information that the audiences want to know. Besides, by the aid of an intelligent information mining algorithm and features extracted by a color co-occurrence matrix, finding of a facial expression recognition area of interest and feature extraction are combined into one step, the faces are clustered by the aid of an affinity propagation clustering algorithm, and the broadcasting information is changed according to emotion. By the aid of the exhibition system, the information can be effectively pushed according to ages and genders of the audiences in a complicated background environment, sufficient exhibition can be realized, the audiences can be guided to more comprehensively know the information, and the identity sense of the audiences is enhanced.

The invention provides an Alzheimer's disease and mild cognitive impairment identification method based on two-dimension features and three-dimension features. The method particularly comprises a step of performing pretreatment of a medical image, wherein the pretreatment comprises pre-segmentation, registration and other processes; a step of performing two-dimension textural feature extraction of the medical image, wherein features comprise the quadratic statistic of a gray-level co-occurrence matrix and a multiscale and multidirectional feature value of Gabor wavelet transformation; a step of performing three-dimension morphological feature extraction of the medical image, i.e., extracting volume features of an area of interest; a step of performing feature fusion of three-dimension morphological features and two-dimension textural features; and a step of constructing a support vector machine to achieve identification of Alzheimer's disease and mild cognitive impairment. According to the method provided by the invention, the three-dimension morphological features and the two-dimension textural features are combined, so that the content of the medical image can be expressed in a comprehensive and accurate manner. The method can improve identification of Alzheimer's disease and mild cognitive impairment, thereby providing a more effective clinic assistant diagnosis.

The invention provides a video target detecting method based on a cascade regression convolutional neural network. The video target detecting method comprises the following steps of 1, inputting a video image sequence, and performing CNN characteristic extraction on all image frames of the whole video sequence through the convolutional neural network; 2, classifying the last convolutional characteristic layer of the CNN characteristic by means of an RPN network for obtaining a suggested area, performing cascade classification and regression on the suggested area through a multiscale convolutional characteristic, and obtaining a static picture detecting result of each image frame; 3, using the results with confidences which are larger than 0.6 in the detecting results that are obtained in the step 2 as initial tracking values, tracking the target on the conv5-3 convolutional characteristic of the CNN characteristic through related filtering for obtaining a time sequence suggested area,performing cascade classification and regression on the time sequence suggested area, and obtaining a time sequence detecting result; and 4, suppressing abnormal values in the static picture detectingresults and the time sequence detecting results through a co-occurrence matrix, thereby obtaining a final detecting result.

A facial expression identification method based on a Gabor wavelet and a gray-level co-occurrence matrix. According to the method, firstly, a pure face region in an expression image is extracted by a manual partitioning method and color image gray processing, histogram equalization and scale normalization processing are carried out; then data redundancies of conventional Gabor features are greatly reduced by a method for carrying out block extraction on a Gabor feature statistic, and the gray-level co-occurrence matrix which reflects the image texture features is introduced into the field of expression identification for the first time so as to make up the defect of deficiency of pixel spatial relativity, which is caused by Gabor feature block processing; and finally, a set of low-dimension feature vectors for feature expression are generated and on the premise of ensuring a high expression identification rate, the expression identification rate is greatly improved.

A cost-effective and robust method for localizing and tracking drowsiness state of the eyes of driver by using images captured by near infrared (IR) camera disposed on the vehicle, the said method comprising the processor implemented steps of: Real-time tracking of the face and localizing eye bounding box within the face bounding box in the captured image by comparing the gray values with threshold using the segmentation process; tracking the eyes by computing the centroid of the eye, target model histogram and target candidate model histogram for one location to another by comparing them to identify distance and calculating the displacement of the target centre by the weighted means, wherein the target model histogram and target candidate model histogram are computed based on the feature space; and detecting the drowsiness state of the eyes using histogram equalization, Morphological operations and texture based parameters using histogram and grey level co-occurrence matrices.

A fake fingerprint detection method based on markov random field (MRF) and support vector machine-k nearest neighbor (SVM-KNN) classification includes steps: (1) fingerprint image feature extracting: (1.1) first-order statistics (FOS), (1.2) a gray level co occurrence matrix (GLCM) and (1.3) an MRF; (2) SVM training: training the FOS and the GLCM feature vector and the MRF feature vector to obtain a model A and a model B; (3) SVM-KNN classification: (3.1) the SVM classification mechanism and (3.2) SVM-KNN classifier forming; and (4) decision fusion for true and false fingerprint detection. Presently related articles for fake fingerprint detection by aid of the GLCM and the MRF are not found, and the fake fingerprint detection method achieves the purpose of identifying true and false fingerprints by aid of physical structures of the two feature quantized fingerprint images. Experiment results prove that the false accept rate and the false reject rate of the algorithm are respectively 1.84% and 1.79%, and therefore the fake fingerprint detection method is high in accuracy and good in practicality.