57 results about "Wavelet fusion" patented technology

The invention relates to an electrocardiogram signal classifying method based on one-dimensional convolutional neural network. The method comprises the following steps: firstly, conducting denoising processing on an electrocardiogram signal by virtue of a wavelet fusion method; then, detecting an R wave peak point by a QRS wave group recognition algorithm based on biorthogonal spline wavelet, and completing division and dimensionality reduction of the electrocardiogram signal with the R point as a datum, so that a plurality of R wave candidate bands are obtained; then, establishing and optimizing an electrocardiogram signal oriented one-dimensional convolutional neural network model; and finally, with the processed R wave candidate bands as input data of the model, automatically completing characteristic extraction and classification of the electrocardiogram signal. The method provided by the invention, with the adoption of the wavelet fusion method, can simultaneously remove high-frequency noise and low-frequency noise, so that extracted signal characteristics are more conducive to recognition; by establishing the electrocardiogram signal oriented one-dimensional convolutional neural network model, the problem that electrocardiogram signal characteristic points must be precisely located is solved, and moreover, the problem that a conventional method, which selects an algorithm to extract characteristics firstly, and then selects an algorithm to complete classification, is complex in computation is solved.

The invention discloses an infrared and visible light image fused automobile anti-blooming video image processing method, solving the traffic safety problem caused by automobile blooming at night from the view of collision object avoidance. The method includes acquiring an infrared image and a visible light image of a night road; performing image registering by taking the infrared image as a reference, enhancing the infrared image and the visible light image; extracting the brightness signal component Y and two chrominance signal components U and V of the enhanced visible light image; fusing Y and the enhanced infrared image wavelet to obtain a new brightness signal component Y'; and subjecting Y' and U and V to YUV inverse transformation to obtain a final color fusion image. The image enhancement processing is added after registration, and the parameters are optimized, the brightness and dark detail information of the fused image are greatly retained and improved while the blooming is eliminated. The method can effectively prevent the collision caused by blooming and is applicable to the automobile anti-blooming filed.

The invention discloses an SAR (Synthetic Aperture Radar) image change detection method based on an adaptive weight and a high frequency threshold, which mainly solves problems that the detection effect is not ideal in the prior art and that a single-type difference chart is low in detection accuracy. The SAR image change detection method comprises the implementation steps of 1, reading in two SAR image I1, I2 which are acquired from the same area at different moments; 2, respectively calculating a mean ratio difference image Dm and a neighborhood log ratio difference image Dl of corresponding pixels of the image I1 and the image I2; 3, carrying out wavelet fusion on the mean ratio difference images Dm and the neighborhood log ratio difference images Dl to acquire fused difference images Xd; and 4, clustering the fused difference images Xd into two different categories so as to acquire a change detection result. The SAR image change detection method is simple to operate, good in noise resistance and high in detection accuracy, can acquire good effects for different types of SAR images, and can be applied to environment monitoring, marine observation, disaster evaluation, resource exploration, urban planning and geographical mapping.

The invention discloses a control method of light-emitting diode (LED) illumination. The method comprises the steps: judging whether a person exists in an LED lighting space or not; when the fact that the person exists in the lighting space, driving a charge coupled device (CCD) sensor to conduct image collection on the lighting space with the person; carrying out pre-processing on multiple images collected in the same lighting space within a certain preset time period; carrying out fusion on the pre-processed multiple images through a wavelet fusion technology to obtain a fusion image; calculating an average gray-value of the fusion image, comparing the average gray-value with a preset standard value, and judging whether the average gray-value of the fusion image exceeds the preset standard value or not; and controlling an LED lamp to adjust the illumination when the average gray-value of the fusion image exceeds the preset standard value. The invention discloses a control system of the LED illumination. The situation that adjustment on the LED illumination is conducted in real time according to change of factors such as actual illumination situations of the lighting space and behavior situations of people is achieved, and energy consumption of the LED lamp is correspondingly reduced under the situation that the enough illumination is guaranteed.

The invention discloses a SAR image change detection method based on a priori, a fusion gray level and a textural feature. By using the method of the invention, problems that a Gaussian model can not completely fit distribution of a difference graph and change detection accuracy is low because only pixel gray level information of the SAR image is used are mainly solved. The method comprises the following realization steps that (1) two time phase SAR images which are registered and corrected are read in; (2) a wavelet fusion strategy is performed on the two images so as to construct the difference graph; (3) a classified priori probability of the difference graph is calculated; (4) the gray level of the difference graph and the texture information are fused so as to acquire an observed quantity likelihood probability; (5) the classified priori probability and the observed quantity likelihood probability are used to calculate a posteriori probability; (6) a maximum posteriori probability criterion is used to divide the difference graph into a change type and a non-change type; (7) a step (3) to a step (6) are repeated till a terminal condition is satisfied and a final change detection result is output. The method of the invention has the advantage that change detection precision to the SAR image is high. The method can be used to extract and acquire change detail information of the SAR image.

The invention discloses an industrial CCD color imaging system based on super-resolution reconstruction and automatic registration of an FPGA and a DSP. The imaging system consists of a hardware system and a software system, wherein the hardware system comprises a video acquisition module which uses a color CCD camera and a monochrome CCD camera to perform real-time acquisition of different resolutions, a module used for analog video digitalization and system conversion, a video pre-processing and automatic and reconstruction parameter registration module based on the FPGA, an automatic registration and trous wavelet fusion reconstruction algorithm realization module taking the DSP as the core, and an Ethernet communication interface module; and the software system is mainly a fusion algorithm which realizes a path of low-resolution color video and a path of high-resolution monochrome video through automatic registration and then adopts a fusion algorithm of combining trous wavelet and IHS to reconstruct a needed high-resolution color video.

The present invention discloses an analog circuit fault diagnosis method based on a chaos cloud model adaptive firefly algorithm. The method comprises the steps of: applying a certain test stimulus toa tested circuit, and collecting output response signals of the tested circuit at a measurable node of the circuit; employing the wavelet fusion method to perform extraction of a circuit fault feature set from the output response signals; employing the CCAFA-LSSVM (Chaos Cloud Model Adaptive Firefly Algorithm-Least Squares Support Veotor Maohine) to perform fault diagnosis of the fault feature set to achieve classification and location operation of the circuit faults. The analog circuit fault diagnosis method employs a cloud model with a capacity of processing object fuzziness and randomnessto perform corresponding improvement of the firefly algorithm and allow the firefly algorithm to have a good generalization ability and a high robustness. Therefore, when the fault diagnosis is performed, the resolution ratio of the fault mode is high, the diagnosis performance is good, and the fault elements can be accurately located, and the diagnosis performance and the efficiency of the testedanalog circuit can be improved.

The invention relates to a multi-wavelet fusion image processing method for polarization haze removal, and belongs to the technical field of image processing. According to the haze removing method, polarization detection is combined with a multi-wavelet fusion algorithm, on the basis of target polarization detection, a plurality of scale functions and a plurality of wavelet functions of multiple wavelets are utilized, properties such as symmetry, orthogonality and compact support are met simultaneously, and different rules are respectively adopted for fusion of polarization information such as the polarization degree and the polarization angel and the intensity information of a target according to characteristics of decomposed high / low frequency coefficients. By using the method, recognized details and brightness information of the target can be simultaneously integrated by a polarization fusion image, a target image with a relatively high recognition rate, relatively high quality and relatively much detail information is obtained in a haze weather environment, and the haze removal property of target detection is improved.

The invention relates to a multi-wavelet fusion feature based neuron action feature extraction method. The method comprises the following steps of: denoising an original action potential signal by using fast wavelet transformation to acquire a denoising action potential signal; performing multi-wavelet base analysis on the denoising signal to acquire a plurality of groups of wavelet time frequency features; respectively fusing the wavelet features with different sizes to acquire a multi-wavelet feature of action potential; and respectively fusing a low-frequency component and a high-frequency component of the signal through the multi-wavelet fusion feature according to features of different wavelet bases to acquire a group of new time frequency features. By adopting the method, mutation and phase step features of the signal are kept, so that the high-frequency component and the low-frequency component of the signal are restored to a certain extent; meanwhile, information and position of a phase step or mutation point of the signal are kept unchanged.

The invention relates to an X-ray image enhancement method based on dual-energy spectrum and wavelet fusion. The method includes the steps of obtaining material specific attenuation changes by changing energy spectrum distribution of an X-ray source in accordance with the principle that attenuation coefficient of a matter is related to energy; selecting the maximum of high frequency coefficient gradients and multiplying the maximum by a weighting factor (the weighting factor is greater than 1) according to the low-contrast feature of X-ray CT (computerized tomography) imaging as a fusion method for increasing high frequency coefficients and highlighting details; calculating the fractal dimension of low frequency coefficients to obtain the complexity, performing image fusion enhancement on dual-energy spectrum two images through objective and subjective weight fusion factors in the maximum entropy principle for overcoming the deficiency that weight selection has no theoretical basis and is too subjective, and making up for deficiencies of a single CT image. The X-ray image enhancement method based on dual-energy spectrum and wavelet fusion is significant.

The invention belongs to the image processing technology field, and particularly relates to an image rain removing method based on a wavelet analysis and a system. The image rain removing method based on wavelet analysis comprises steps of performing wavelet multilayer decomposition on a video frame image according to wavelet analysis, analyzing image information of all image layers and detecting the image layer containing rain droplet noise, c determining a rain droplet contamination degree coefficient according to the brightness characteristic of the rain droplet, calculating a fusion coefficient matrix according to the rain droplet contamination degree coefficient, performing wavelet fusion on the image layer containing various image information according to the fusion coefficient matrix, and performing image reconstruction through the fusion result to obtain the rain removing image. The invention can avoid the interference of the dynamic characteristic, can more effectively and accurately remove rain droplet, can improve the usage scope of the algorithm, enables the algorithm to have a sound rain-removing effect even under the heavy rain, performs defogging operation on the rain removing image, optimizes the visual effect of the image and enables the algorithm to be more practical.

The invention relates to the technical field of underwater image processing, and provides an underwater image enhancement algorithm under a low illumination condition. Therefore, the technical schemeadopted by the invention is as follows: the low-illumination underwater image enhancement method based on multi-scale detail enhancement comprises the following steps: firstly, removing color cast ofan image by improving a color correction coefficient of a gray world algorithm, so that the image has a very good visual effect; the introduced multi-scale detail enhancement algorithm enhancing the detail information of the image, and finally wavelet fusion being performed on the color-corrected image and the contrast-enhanced image so as to further perform noise reduction and enhancement on theimage. The method is mainly applied to underwater image processing occasions.

The invention discloses a method for improving image segmentation effects by using a wavelet fusion algorithm. Through the novel method that after image segmentation is carried out on images through an OTSU algorithm and a two-peak method respectively, wavelet fusion is carried out on image segmentation results obtained through the OTSU algorithm and the two-peak method by using the wavelet fusion algorithm, target areas obtained through the segmentation are made to be more complete and clearer in detail.

The invention belongs to the image processing technical field and relates to a wavelet analysis-based image rain-removing method and a wavelet analysis-based image rain-removing system. The wavelet analysis-based image rain-removing method includes the following steps that: image layer decomposition is carried out on a video frame image according to wavelet analysis, and the image information of all decomposition image layers are analyzed; step b, a fusion coefficient matrix is calculated, wavelet fusion is carried out on decomposition image layers with different image information according to the fusion coefficient matrix, and image reconstruction is carried out according to a fusion result, so that a rain removed image can be obtained; and step c, filtering processing is performed on the rain removed image by using anisotropic diffusion filtering. With the wavelet analysis-based image rain-removing method and system of the invention adopted, the interference of dynamic characteristics can be avoided, rain droplets can be removed more effectively and accurately, the usage scope of the rain removing algorithm can be extended, the algorithm can have a sound rain-removing effect even under heavy rain; filtering processing is performed on the rain removed image through the anisotropic diffusion filtering, so that the image can be clearer and more nature, and the rain removing algorithm can be more practical.

The invention relates to a multi-mode image processing method and a multi-mode image processing platform. The method comprises the steps of reading SPECT data, PET data and CT data; preprocessing theread data; displaying the data in a three-dimensional mode; segmenting a selected region in an image by adopting a segmentation algorithm; correcting the image according to a long axis and a short axis of a heart in the image; taking a CT image as a template by adopting a mutual information method, registering PET and SPECT images to the CT image, and obtaining registered SPECT and PET images; fusing registered SPECT and PET images in a wavelet fusion mode; displaying a bullet graph of SPECT, PET and fused images; carrying out the operations of measurement, polygonal drawing, zooming, rotating and / or marking on the obtained bullet graph, performing quantitative analysis on an obtained result, and providing quantitative diagnosis indexes for doctors based on the difference among the bullet graphs of different modal data. According to the invention, various image data can be read. Through the operations of matching, segmenting, fusing, three-dimensional visualization and the like, image data are analyzed and processed. Therefore, the usage is facilitated, and the efficiency is remarkably improved.

The invention provides a method and a system for processing the bright and dark faces of a face image. The method comprises the following steps: acquiring a first face image and judging the bright anddark faces of the first face image; if the first face image is judged to have a bright and dark face, starting the method of combining wavelet illumination compensation and LBP to modify the first face image to obtain a second face image; calculating an LBP histogram of a rotation-inconvenient uniform mode for the first face image, obtaining an LBP histogram of the first face image, and applyingthe LBP histogram of the first face image to the second face image to obtain a third face image; performing wavelet fusion of the second face image and the third face image to obtain a fourth face image. The invention solves the technical problem that the existing technology of face image recognition cannot effectively remove the bright and dark faces and cannot effectively complete the correction of the bright and dark faces.

The invention provides an SAR image change detection method based on directional wavelet transformation and an improved level set. The SAR image change detection method comprises the achieving steps of 1 inputting SAR images A and B to be preprocessed; 2 adopting different ratio methods to establish ratio difference charts by using preprocessed images; 3 fusing the ratio difference charts based on the directional wavelet transformation; 4 segmenting the fused ratio difference chart; 5 obtaining a change detection result chart. Directional wavelet fusion is performed on a logarithm ratio difference chart and an average ratio difference chart, so that background information is effectively inhibited, change information is strengthened, and accordingly a change area and an unchanged area are accurately described. In addition, due to the fact that an improved SBGFRLS level set model is adopted to perform image segmentation on fused difference images in the method, the accuracy of classification on the change information and non-change information is improved, a change detection result can be accurate, and meanwhile the complexity of change detection time is remarkably reduced.

The invention discloses an image splicing and fusing method and system. The method comprises the following steps: acquiring an image to be spliced; establishing an image scale space to obtain an imageof each image in m different scale spaces; solving feature points of the image in all scale spaces, and solving an optimal feature point through a non-maximum suppression method; solving the direction of the optimal feature point; establishing a feature descriptor, and performing feature point matching by utilizing the feature descriptor; acquiring an image registration mapping matrix, and completing registration of the two images; and after the registration of the two images is completed, fusing the overlapping regions of the two images to complete image splicing. The feature point solving method and the feature point descriptor solving method are high in speed, small in calculated amount, good in robustness and insensitive to geometric deformation. The image registration method is highin precision, high in reliability and high in anti-interference performance. Based on wavelet fusion, two spliced images can be better transited, information is not lost, and detail information is restored to the maximum extent.

The invention provides a method for processing footprint by multi-angle acquisition and integrated extraction. The flash light is used to obtain multiple footprint images with dark field illumination and background image with bright field illumination. High frequency information of each image in multiple footprint images is fused through wavelet fusion method to reduce the loss of footprint image details. Then the fused footprint images and background images can be handled respectively through the lightning correction method based on the block to solve uneven illumination. Subtraction can be made between the footprint image and background image after lightning correction, which can not only reduce noise but also extract the footprint texture, edge and other significant information more effectively, leading the effective area of footprint image more complete and the details more clearly.

The invention discloses a fast wavelet fusion method for fast moving object thermal imaging image and optical image, which includes preprocessing thermal imaging image and optical image, HSI decomposing thermal imaging image and optical image, wavelet decomposition processing, wavelet fusion processing and wavelet reconstruction processing to obtain fused image. As that thermal image and the optical image are HSI decomposed, The decomposition layers are fused by wavelet fusion method, which can effectively synthesize the thermal imaging image target characteristic information and optical imagescene details information, and obtain more comprehensive information fusion image, so as to realize the effective imaging of fast moving objects.

A kidney multiperiod CT image fusion system disclosed by the present invention comprises an image acquisition device used for obtaining the kidney CT images, a pre-processing device for denoising andenhancing the kidney CT images, a primary registration device for grouping the kidney CT images, a multi-phase fusion device for fusing the kidney CT images to realize the information complementationbetween the kidney CT images of different phases, a three dimensional reconstruction device for generating the specific kidney three dimensional images based on the fused images and a kidney databasefor classifying and storing the specific kidney three dimensional images. The multi-phase fusion device comprises a linear weighting fusion unit, a high-pass filtering fusion unit and a wavelet fusionunit, the linear weighting fusion unit multiplies the corresponding pixel values in the kidney CT images by a weight coefficient separately and then adds together to obtain the fused images, the high-pass filtering fusion unit superposes the high-frequency components of a high-resolution image to a low- resolution image to obtained the fused images, and the wavelet fusion unit carries out the wavelet decomposition and then the inverse wavelets transform on the kidney CT images to obtain the fused images.

The invention discloses a multi-scale Gabor wavelet PCA fusion image edge detection method, and the method specifically comprises the steps of: 1, converting a to-be-detected color image into an LMN color space to obtain an L channel image, an M channel image and an N channel image; 2, constructing a multi-scale and multi-direction Gabor imaginary part filter bank, and enabling the three channel images to respectively pass through the imaginary part filter bank to obtain a filtering response; 3, for the filtering response of each channel component, adopting a PCA method to fuse the filtered images of the same scale in different directions, and calculating the gradient of each channel; and 4, carrying out gradient fusion on all the components, calculating the amplitude and direction of thegradient, and carrying out non-maximum suppression and a double-threshold method to obtain the edge of the color image. The problem that in the prior art, due to the fact that gray level image structure information obtained through dimension reduction is incomplete, image edge information is lost is solved.

The invention provides a small target enhancing detection method based on secondary image fusion. The method comprises the following steps of: dividing eight continuous frames of infrared image sequences into two groups, wherein the first four frames belong to a group A, and the last four frames belong to a group B; carrying out fusion on the images of the group A and the group B by a weighted average fusion algorithm to obtain two primary-fusion images; independently carrying out wavelet decomposition on the two primary-fusion images to obtain the high-frequency part and the low-frequency part of each image; adopting different fusion criterions to carry out respective fusion processing on the high-frequency part and the low-frequency part in the step 3; and carrying out wavelet reconstruction on the fused high-frequency part and low-frequency part to obtain a secondary-fusion image. By use of the method, a secondary image fusion algorithm is introduced into the field of infrared dim target detection, the target in a continuous frame is regarded as a stationary state, the primary fusion is firstly carried out, stochastic noise in the image is inhibited, a fusion result is subjected to multi-scale wavelet fusion, a dim target is enhanced while background is inhibited, and a target detection probability is improved.

The invention relates to the technical field of fabric automatic classification, and provides a fabric abnormal texture type identification method and device, and the method comprises the steps: enhancing the quality of an original abnormal texture image through a preset image enhancement model, and generating an enhanced abnormal texture image; performing multi-level multi-direction decompositionon the enhanced abnormal texture image through a preset dual-tree complex wavelet decomposition model to obtain a frequency domain texture feature set; extracting 6*N high-frequency abnormal texturesub-graphs from the frequency domain texture feature set, fusing the six high-frequency abnormal texture sub-images in each level into a high-frequency abnormal texture fusion image of the corresponding level through a preset dual-tree complex wavelet fusion model; and identifying the abnormal texture type to which the fabric belongs according to all high-frequency abnormal texture fusion images.Thus, adaptive and stable transformation in different levels and different directions is promoted to obtain a frequency domain texture feature set, and accurate and efficient identification of the abnormal texture type of the fabric according to the frequency domain texture feature set is well supported.

The invention discloses an image enhancement method and system, and the method comprises the steps: S1, carrying out wavelet decomposition of a collected foggy image, and obtaining a low-frequency component; S2, carrying out wavelet reconstruction on the low-frequency components, so that low-frequency sub-images can be obtained; S3, estimating an atmospheric dissipation function, an atmospheric light value and medium transmittance based on the low-frequency sub-images; S4, generating a defogging low-frequency sub-image based on the atmospheric dissipation function, the atmospheric light valueand the medium transmittance; S5, obtaining a first defogged image of the foggy image by adopting a dark channel prior defogging method; S6, performing wavelet decomposition on a first defogged imageto obtain a high-frequency component; S7, performing wavelet reconstruction on the high-frequency component to obtain a high-frequency sub-image corresponding to the first defogged image; and S8, performing wavelet fusion on the low-frequency sub-image and the high-frequency sub-image to generate a second defogged image. The second defogged image generated by the invention can realize detail enhancement at the same time, and the brightness change effect of images is natural.

The invention provides a water quality analyzing method based on signal processing and discloses a water quality analyzing method through the processing of water quality monitoring data in the signal processing and environment field. The method comprises three parts: 1) decomposing a signal based on a sparse linear equation group; 2) further decomposing the structural parts of the signal obtained from the last step; and based on the first part's signal decomposition method and the adding method of the wavelet fusion method, integrating the high-resolution signal and the multi-spectral signal to obtain a new signal. The method of the invention utilizes the signal decomposing and integrating technology to realize the full-automatic water quality analysis based on signal processing.

The invention discloses a seismic data event detection method of an environment suppression and contour combination model based on visual cognition. The method comprises steps that an event is picked up based on pre-stack seismic data which contains abundant information such as amplitude and frequency, and can reflect a tiny structure of a stratum; firstly, seismic data are preprocessed, wavelet fusion of envelope peak instantaneous frequency and oblique superposition peak instantaneous amplitude of the seismic data is solved, and optimal mass data can be provided to the maximum extent. While the seismic data event contour is enhanced, anisotropy suppression is carried out on environment non-correlation noise, significant extraction of the event target contour is realized, and preparation is made for subsequent seismic interpretation, reservoir prediction and the like. Compared with an isotropic method, the anisotropic environment suppression method adopted by the invention can more accurately detect edge variation.

The invention discloses a control method of light-emitting diode (LED) illumination. The method comprises the steps: judging whether a person exists in an LED lighting space or not; when the fact that the person exists in the lighting space, driving a charge coupled device (CCD) sensor to conduct image collection on the lighting space with the person; carrying out pre-processing on multiple images collected in the same lighting space within a certain preset time period; carrying out fusion on the pre-processed multiple images through a wavelet fusion technology to obtain a fusion image; calculating an average gray-value of the fusion image, comparing the average gray-value with a preset standard value, and judging whether the average gray-value of the fusion image exceeds the preset standard value or not; and controlling an LED lamp to adjust the illumination when the average gray-value of the fusion image exceeds the preset standard value. The invention discloses a control system of the LED illumination. The situation that adjustment on the LED illumination is conducted in real time according to change of factors such as actual illumination situations of the lighting space and behavior situations of people is achieved, and energy consumption of the LED lamp is correspondingly reduced under the situation that the enough illumination is guaranteed.

The invention discloses an optimized fusion method based on fuzzy integral, in the HIS space, the intensity component of the spectrum image is decomposed with wavelet multi-layer and acquires the low frequency base band coefficients, acquires the low frequency base band coefficient by decomposing the high resolution image with multi-layer wavelet, integrates the two single factor indexes of spectrum information and space resolution are synthesized with fuzzy integral, carries on pixel fusion with iteration, and carries on character fusion to the high frequency detail of the high frequency sub-band coefficient after being decomposed with wavelet, then carries on correspondent wavelet reverse conversion to the sub-band coefficient and low frequency coefficient, and acquires the new intensity component I', then the fused image may be acquired through HIS reverse conversion. The invention combines the character of HIS fusion method and wavelet fusion method, it realizes high space resolution, and it reduces the deformation of color greatly, and it improves the spectrum information index of fused image.