94 results about "Normalized correlation coefficient" patented technology

The invention discloses a real-time visual target tracking method based on light streams. The method includes: screening feature points rich in textural features, and tracking the feature points between every two frames of images to obtain feature point matching relations; filtering the feature points with large matching errors through various filter algorithms such as normalized correlation coefficients, forward-reverse tracking errors and random consistency detection to retain most reliable feature points so as to obtain the target motion speed in images; using the target motion speed as observation quantity, and estimating through Kalman filtering to generate the target position. Compared with a tracking algorithm, the method has the advantages that the method is unaffected by light, the target can be stably tracked under the conditions that the target moves, a camera moves, the target is blocked for a short term and the like. The method is high in calculation efficiency, and the tracking algorithm is verified for the first time on an ARM platform.

The invention discloses a medical image robust zero-watermarking method based on Bandlet-DCT, and the method comprises the steps: carrying out the Bandlet-DCT of an original medical image, obtaining acoefficient feature matrix, and generating a feature binary sequence through Hash function operation; carrying out chaos scrambling encryption on the original watermark to obtain a chaos scrambling watermark, embedding watermark information into the original medical image, and acquiring and storing a binary logic key sequence in a third party; similarly, generating a feature binary sequence of the to-be-detected medical image; extracting an encrypted watermark according to the characteristic binary sequence and the binary logic key sequence, and decrypting the encrypted watermark to obtain areduced watermark; and performing normalization correlation coefficient calculation on the original watermark and the reduced watermark, and determining the ownership of the medical image to be testedand the embedded watermark information. The method has good robustness and invisibility in geometric attack resistance and conventional attack resistance, and can protect the privacy information of apatient and the data security of a medical image at the same time.

The invention relates to an intelligent-texture anti-counterfeiting method based on perceptual hashing. A first step is characterized in that image characteristic extraction is performed, which comprises that (1) a perceptual hashing algorithm is used to process an image so as to obtain one visual characteristic vector V (j) of an original texture image; (2) a user uses a mobile phone to scan the texture image to be tested and upload to a server, the perceptual hashing algorithm is used to process the image to be tested and the visual characteristic vector V ' (j) of the image to be tested is acquired. A second step is characterized in that image discrimination is performed, which comprises (3) a normalized correlation coefficient NC value between the visual characteristic vector V (j) of the original texture image and the visual characteristic vector V ' (j) of the image to be tested is acquired; (4) the obtained NC value is returned to the mobile phone of the user. An experiment proves that the method of the invention possesses a strong conventional attack resistance capability and a geometric attack resistance capability. A problem of automatically discriminating the texture image is solved. An intelligent texture anti-counterfeiting technology is realized. Discrimination accuracy is high and a speed is fast.

An image defect inspecting apparatus of the present invention includes a template image producing section for producing a template image from reference image data, a corresponding image extracting section for extracting a predetermined image located at a position corresponding to a template image from digital data of a scanned image for inspection, data embedding sections for embedding desired same pattern data into the template image and the image extracted by the corresponding image extracting section, a normalized correlation value calculation processing unit for acquiring a normalized correlation coefficient from the template image and the extracted image, into which the pattern data is embedded, and a defect judging section for judging as to whether a defect is present by comparing the normalized correlation coefficient acquired by the normalized correlation value calculation processing unit with a predetermined threshold value so as to acquire a large/small relationship thereof.

The invention provides an automatic voice harmonic-to-noise ratio analysis method, which comprises: 1) an effective phonetic segment for harmonic-to-noise ratio analysis is segmented from the recording; 2) the phonetic segment is subject to filtering processing based on an auditory model, and then a two-dimensional energy correlation coefficient of a time domain and a frequency domain in various filter channels in the auditory model is calculated; and 3) the threshold value of the correlation coefficient is preset, a coordinate point of the time domain and the frequency domain corresponding to the correlation coefficient has a harmonic component when the correlation coefficient obtained in step 2) is larger than the threshold, or else the coordinate point of the time domain and the frequency domain corresponding to the correlation coefficient has a noise component, and finally the ratio of the harmonic component and the noise component is calculated to obtain the harmonic-to-noise ratio. The method uses the correlation between the time domain characterized by an autocorrelogram and a cochlear spectral domain channel to judge the harmonic component, is not affected by the detection position of a fundamental frequency, and can detect the harmonic component more accurately and more robustly. As the cochlear spectrum is used, the method is more matched with the actual hearing of human ears.

It is an aim image automatic identifying and rapid tracing method, which comprises the following steps: first to establish aim image valve grey scale grade; second to establish error and correlation index; then to form mode image and spot image characteristics matrix on the base of valve value grey scale and to form correlation functions under the condition of discreteness, such as variance, normalization correlation index, inner product operation in the linear space of all two-dimension matrix; to get the correlation definition according to norm and angle in the Euclidean space starting from the linear space; then to get a judge formula by use of inner product definition to realize the rapid identification of the aim.

The invention relates to a method for generating digital respiratory gating signals based on an abdominal body surface outline, which comprises the following steps: (1) respectively extracting the abdominal body surface outline in each frame image of a reference abdominal video, and calculating normalized correlation coefficients NCC of the abdominal body surface outline at each moment and a reference outline to obtain an NCC-time curve; (2) arranging a grating window on the NCC-time curve, and providing a grating type label for the each frame image; (3) converting the each frame image of the reference abdominal video into a column vector to form a training set containing the type label; (4) reducing the dimension of the training set to obtain a support vector omega and a classification threshold value b by training; and (5) converting a real-time abdominal video into a column vector, and using the support vector omega and the classification threshold value b to calculate a sorting signal of each column vector to obtain sorting signals, namely, the gating signals. The method can lower cost, cannot bring uncomfortable feeling for a patient, and has good tolerance.

The invention discloses a color image total-blindness robustness digital watermarking method based on a self-embedding technology. The method includes: performing discrete Fourier transform on each sub-block in an approaching sub-graph of a brightness component of a color image to obtain an amplitude spectrum matrix of a discrete Fourier transform coefficient matrix of each sub-block; creating a characteristic watermark and a self-embedding characteristic watermark according to the amplitude spectrum matrixes of all the sub-blocks; obtaining an amplitude spectrum matrix of a discrete Fourier transform coefficient matrix of each sub-block in an approaching sub-graph of a brightness component of a watermarking color image in the same mode; performing blind extraction of a characteristic watermark and an authentication watermark according to the amplitude spectrum matrixes of all the sub-blocks; and performing copyright protection according to a normalization correlation coefficient between the characteristic watermark and the authentication watermark obtained through blind extraction. The method is advantageous in that the embedding intensity of a digital watermark has the adaptability for an original color image and has ideal robustness for image processing attacks such as geometric translation etc., and any information of the original color image and the original digital watermark is avoided during extraction of the digital watermark.

The invention provides a mutual inductor infrared image recognition method based on image textual features. The method includes the steps of conducting graying and denoising on an image, and conducting the X-direction Canny detection on the image, wherein the textural features of the left edge an the right edge of a sleeve are exposed after graying, denoising and detection are conducted; fitting left feature points into a straight line and fitting right feature points into a straight line after the methods of the normalized correlation coefficient maximum value matching and K-means clustering are conducted, and determining the highest edge points and the lowest edge points on the straight lines so as to determine the image area of a mutual inductor. By means of the method, the defect that a traditional image recognition method is not enough in accuracy is overcome, the background interference prevention capacity is high, the specific area of the mutual inductor in the infrared thermal image can be rapidly recognized, operating personnel are helped to detect the position coordinates of the mutual inductor, the automation degree of the state monitoring of a power system is improved, and the very high practical value is achieved.

The invention discloses an efficient realization method for normalized correlation image template matching, and belongs to the technical field of image processing. The matching method is realized by constructing an apparatus comprising a high-speed correlation operation module, a template drawing and real-time graph data external memory, a microprocessor and a result memory. Normalized correlation coefficients of template drawings and real-time graphs in all columns and a current row in a search position are concurrently computed in a concurrent processing mode at first, template drawing data are re-ordered according to a row sequence of real-time graph data during next-row search computing, and then normalized correlation coefficients of template drawings and real-time graphs in all columns and a next row in the search position are concurrently computed, so that normalized correlation coefficients of template drawings and real-time graphs in all the columns and all the rows in the search position can be computed on the premise that the real-time graph data correspond to the template drawing data.

The invention relates to an cell in-situ active deformation measurement method, which comprises the following measurement steps: (1) dividing square sampling grids on a cell-base fluorescent reference image and a corresponding deformation image; (2) determining the sizes of a reference subarea and a searching area; (3) establishing four global summation table; (4) calculating the cross correlation coefficient between the image reference subarea and the corresponding searching area, which take a sampling point as a center, by recurrence; (5) quickly calculating a zero-mean normalization related coefficient matrix by looking up the summation table according to the result of the step (4), and determining the overall pixel displacement of the sampling point; (6), performing subpixel interpolation by using a gradient method; and (7) recurring and calculating the displacement of the sampling point to obtain an accurate displacement field of a whole deformed image. The method has the advantages of accurate and high-efficiency measurement and high time-space resolution. The method can be used in cell in-situ active deformation measurement with high time-space resolution.

The invention provides an image sharp degree distribution-based characteristic point matching method. According to the method, an edge detection operator is used for calculating sharp degrees of characteristic points of two images to define a characteristic point description method, and all the characteristic points are matched mainly by calculating normalization related coefficients of the characteristic points in the two images to sequentially perform preliminary matching on the characteristic points, delete false characteristic point matching and calculate a perspective projection matrix of the images, so that the matching accuracy and efficiency are improved.

The invention discloses a medical image robust watermarking method based on Gabor-DCT. The method comprises the steps that feature extraction is conducted on an original medical image through Gabor-DCT transformation, and a feature sequence is generated through a hash function; performing chaos scrambling encryption on the original watermark to obtain a chaos scrambling watermark, embedding watermark information into the original medical image, and meanwhile obtaining a binary logic secret key sequence and storing; similarly, performing feature extraction on the medical image to be detected through Gabor-DCT transformation, and generating a feature sequence; extracting an encrypted watermark according to the feature sequence and the binary logic key sequence, and decrypting the encrypted watermark to obtain a reduced watermark; and performing normalization correlation coefficient calculation on the original watermark and the reduced watermark, and determining the ownership and watermark information of the medical image to be detected. The zero watermark embedded in the method has invisibility and robustness, and the privacy information of the patient and the data security of the medical image can be protected.

The invention relates to an image fusion method based on a wavelet direction correlation coefficient. An image is decomposed into a three-layer low-frequency sub band part and a three-layer high-frequency sub band part by using db3 wavelet conversion, wherein low-frequency sub bands employ a pixel point fusion rule determined based on cyclic shift sub block space frequency correlation coefficient; and for all high-frequency sub bands, high-frequency coefficients are determined by regional energy and a gradient normalization correlation coefficient difference based on direction characteristics of the sub bands. According to the image fusion method designed by the invention, correlation of wavelet coefficient spatial frequencies as well as direction correlation between the energy and gradient is taken into consideration and the coefficient participating in fusion is more important and accurate for subjective and objective quality of the fused image. The fusion precision and practicability are high.

The invention discloses a PHTs-DCT-based medical image robust watermarking method, which comprises the steps of performing feature extraction on an original medical image through PHTs-DCT transformation, and generating a feature sequence by utilizing hash function operation; carrying out chaos scrambling encryption on the original watermark to obtain a chaos scrambling watermark, embedding watermark information into the original medical image, and obtaining and storing a binary logic secret key sequence; similarly, carrying out feature extraction on the medical image to be detected through PHTs-DCT transformation, and generating a feature sequence; extracting and decrypting the encrypted watermark according to the feature sequence and the binary logic key sequence of the to-be-detected medical image to obtain a reduced watermark; and performing normalization correlation coefficient calculation on the original watermark and the reduced watermark, and determining the ownership and watermark information of the medical image to be detected. The method has strong robustness and invisibility, and can protect the privacy information of the patient and the data security of the medical image at the same time.

The invention discloses a non-linear oscillation detection method based on improved adaptive frequency modulation mode decomposition. The method comprises the steps of: (1) collecting a loop output signal of an industrial process to be detected; (2) decomposing the signal by using an improved adaptive frequency modulation mode decomposition method; (3) calculating a mean value of the instantaneousfrequency of each decomposition mode; (4) using the mean value of the instantaneous frequency of the mode with the largest normalized correlation coefficient as the basic frequency; (5) calculating aconfidence interval upper limit and a confidence interval lower limit of all other modules in addition to the mode corresponding to the current basic frequency; and (6) judging whether there is an integer multiple of the fundamental frequency in the confidence interval, if so, deeming that the oscillation belongs to non-linear oscillation; and if not, using the mean value of the instantaneous frequency of the mode with the second largest normalized correlation coefficient as the basic frequency, and repeating the steps (4) to (6). By using the non-linear oscillation detection method disclosedby the invention, the accuracy and reliability of non-linear detection of the control loop of the industrial process can be improved.

The invention discloses an intelligent texture anti-counterfeiting method based on DWT-DCT (Dreamweaver Template-Discrete Cosine Transform) transformation, belonging to the technical field of texture anti-counterfeiting. The intelligent texture anti-counterfeiting method comprises the following steps of: firstly establishing a feature database, to be specific, (1) carrying out wavelet transformation on texture images and then carrying out full-graph DCT transformation on approximate sub-images, thus extracting a feature vector V(n), and (2) storing the determined feature vectors in the textural feature database; and automatically identifying the images, to be specific, (3) scanning texture label images to be tested by using a mobile phone, determining the feature vectors V' of the images to be detected by using the method of the step (1), and uploading the feature vectors V' to a server; (4) determining a normalized correlation coefficient NC (N) value between the feature vectors V(n) of all the texture images in the feature database and the feature vectors V' of the images to be detected, and (5) returning the maximum value of the NC(n) to the mobile phone of a user. Experiments prove that the intelligent texture anti-counterfeiting method has the capacity of automatically identifying the texture images, and the intelligent texture anti-counterfeiting technology is realized.

The invention relates to a rail fastener positioning method based on linear template matching. The rail fastener positioning method comprises the following steps: (1) a fastener template is manufactured; (2) according to the width size wf of the fastener area and the coordinates of the steel rail area, and a line-shaped segmentation area of the fastener area is segmented from the original image ofa track line to obtain a line-shaped segmentation area graph, wherein the unit of wf is a pixel; (3) accurate positioning is carried out, a symmetric fastener template on the line-shaped segmentationregion graph is scanned, and the similarity between a scanning area and a template are measured by using a normalized correlation coefficient; and (4) the processes in steps (2)-(3) are recycled, andall the collected rail line original drawings are traversed to position all the fasteners on the track, wherein each track line original drawing only comprises a pair of complete fasteners. The positioning method is suitable for a ballasted line and a ballastless line at the same time, the positioning accuracy is high, the fastener in the damaged state can be effectively positioned, and meanwhilethe positioning speed is 0.3 s/piece or below.

The invention discloses a multivariable correction characteristic wavelength selection method based on a minimum correlation coefficient, and aims to solve the problem of an existing wavelength selection method. The method comprises the following steps: performing S-G first-order derivative processing on the spectral data set X; calculating absolute values of correlation coefficients among the column vectors; obtaining correlation coefficient matrix R, calculating an average value and a standard deviation of other elements except diagonal lines in each column in the correlation coefficient matrix R; selecting a correlation coefficient average value and a standard deviation threshold value pair; forming a to-be-selected wavelength set S; sorting the wavelengths of the S set to obtain a setS'; gradually adding a wavelength variable to establish an MLR model; calculating the RMSEV value of each model, taking the variable subset corresponding to the minimum RMSEV value as the characteristic wavelength under S, selecting the next threshold pair, repeating the above steps, and finding the corresponding minimum RMSEV value and the corresponding characteristic wavelength under all characteristic wavelength sets. According to the variable selection method, redundancy is reduced to the maximum extent, the principle is simple, and implementation is easy.

The invention discloses a medical image robust watermarking method based on Tetrolet-DCT. The medical image robust watermarking method comprises the following steps: performing feature extraction on an original medical image through Tetrolet-DCT and generating a visual feature sequence by utilizing a hash function; performing chaos scrambling encryption on the original watermark to obtain a chaosscrambling watermark, embedding watermark information into the original medical image, and meanwhile obtaining a binary logic secret key sequence and storing; similarly, performing feature extractionon the medical image to be detected through Tetrolet-DCT transformation, and generating a visual feature sequence; extracting an encrypted watermark according to the visual feature sequence and the binary logic key sequence, and decrypting the encrypted watermark to obtain a reduced watermark; and performing normalization correlation coefficient calculation on the original watermark and the reduced watermark, and determining the ownership and watermark information of the medical image to be detected. The zero watermark embedded in the method has invisibility and robustness, and the privacy information of the patient and the data security of the medical image can be protected.

The invention discloses an intelligent texture anti-counterfeiting method based on DCT (Discrete Cosine Transform) transformation, belonging to the texture anti-counterfeiting field. The intelligent texture anti-counterfeiting method comprises the following steps of: firstly establishing a textural feature database, to be specific, (1) carrying out full-graph DCT transformation on each original texture label image, and obtaining feature vectors V(n) in a transfer domain, and (2) storing the determined N feature vectors in the textural feature database; and then automatically identifying the images, to be specific, (3) scanning texture label images to be tested by using a mobile phone, determining the visual feature vectors V' of the images to be detected by using the method of the step (1), and uploading the visual feature vectors V' to a server; (4) determining a normalized correlation coefficient NC (N) value between the feature vectors of all the texture images in the feature database and the visual feature vectors V' of the images to be detected, and (5) returning the maximum value of the NC(n) to the mobile phone of a user. Experiments prove that the intelligent texture anti-counterfeiting method has the capacity of automatically identifying the texture images and the network transmission speed is fast.