35results about How to "Good edge retention" patented technology

The invention discloses an SAR (Synthetic Aperture Radar) image change detection method based on a neighborhood logarithm specific value and anisotropic diffusion, relating to the field of remote sensing image processing and mainly solving the problem that a difference graph structure of SAR image change detection is seriously influenced by SAR image spot noises. The SAR image change detection method comprises the following steps: (1) structuring a difference striograph IL of two images I1 and I2 of different times and same terrain according to a neighborhood logarithm specific value method; (2) carrying out self-adaptation window anisotropic diffusion filtering processing on the difference striograph IL to obtain a final filtering result graph NI<t>[L] of the difference striograph; and (3) carrying out threshold segmentation on the final filtering result graph NI<t>[L] of the difference striograph by using an OSTU (Maximum Between-Class Variance) threshold algorithm to obtain a change detection result graph CNI<t>[L] for structuring the difference striograph by using the neighborhood logarithm specific value method. The histogram of the difference striograph can be compressed so as to effectively eliminate miscellaneous points in the change detection result graph; and the self-adaptation window anisotropic diffusion filtering has favorable edge retentiveness and cannot blurs the edges of the image, thus, an obtained change detection result graph is finer.

The invention discloses an SAR (Synthetic Aperture Radar) image change detection method based on a neighborhood logarithm specific value and anisotropic diffusion, relating to the field of remote sensing image processing and mainly solving the problem that a difference graph structure of SAR image change detection is seriously influenced by SAR image spot noises. The SAR image change detection method comprises the following steps: (1) structuring a difference striograph IL of two images I1 and I2 of different times and same terrain according to a neighborhood logarithm specific value method; (2) carrying out self-adaptation window anisotropic diffusion filtering processing on the difference striograph IL to obtain a final filtering result graph NI<t>[L] of the difference striograph; and (3) carrying out threshold segmentation on the final filtering result graph NI<t>[L] of the difference striograph by using an OSTU (Maximum Between-Class Variance) threshold algorithm to obtain a change detection result graph CNI<t>[L] for structuring the difference striograph by using the neighborhood logarithm specific value method. The histogram of the difference striograph can be compressed so as to effectively eliminate miscellaneous points in the change detection result graph; and the self-adaptation window anisotropic diffusion filtering has favorable edge retentiveness and cannot blurs the edges of the image, thus, an obtained change detection result graph is finer.

The present invention relates to a multi-frequency phase-sensitive demodulation method based on improved median filtering. The method comprises the following steps of: obtaining a digital quantity ofdetected multi-frequency signals through A / D sampling, and performing temporary storage of the digital quantity of the detected multi-frequency signals in a cache module; according to the periodicityof the detected multi-frequency signals, converting the cached digital signals to an M*Nc signal matrix, wherein M is the quantity of the signal sampling periods, and Nc is the quantity of the sampling points in one sampling period; taking each row of elements of the signal matrix, wherein each row of elements forms one M*1 vector to construct Nc rows of vectors; performing median filtering of each row of vectors to obtain the signal vectors after filtering; and performing digital phase-sensitive demodulation according to the signal vectors after filtering to obtain the amplitude and phase information of each frequency component signal.

The invention provides an image resampling method relevant with content and an image resampling system relevant with the content and relates to the technical field of computer image processing. An existing resampling method is usually relevant with a position of a sampling point and lacks of consideration about content of a sampling portion so that a result image has quality deteriorating problems such as mosaics on the image and image vagueness. With the image sampling method, a basic sampling value is obtained with a regular interpolation algorithm, according to adjacent content of a to-be-sampled portion of an original image, the basic sampling value undergoes weighted correction and a final sampling result is synthetised. The problem of vague images in the existing method and the system is avoided with the image resampling method and the image resampling system, and at the same time, high fidelity can be ensured. The image resampling method and the image resampling system can be suitable for application such as image rotation, displacement, abstraction and the like.

An improved non-local mean filtering method for PolSAR images. This method combines the statistical characteristics of the image sub-block and the polarization scattering characteristics of the target point to screen the homogeneous pixels, introduces the coefficient of variation to adaptively select the smoothing coefficient to calculate the weight required for filtering, and performs non-local mean filtering on the homogeneous pixels, and finally Output the filtered PolSAR image. The inventive method has the following advantages: 1) it can effectively filter out coherent speckle noise; 2) it has a strong degree of self-adaptation; the inventive method can adaptively select the smoothing coefficient according to the heterogeneity of the image block; , the edge features of the image are better preserved. 3) The polarization scattering features of the filtered image point target are better preserved by using polarization screening.

The invention discloses a method for carrying out change detection on remote sensing images based on treelet fusion and level set segmentation, and mainly solves the problem that much pseudo-change information exists in the existing change detection methods. The method is implemented through the following steps: inputting two time-phase remote sensing images, then respectively carrying out mean shift filtering on each image so as to obtain two time-phase filtered images; respectively carrying out two-dimensional stationary wavelet decomposition on the two time-phase filtered images three times under different level numbers; carrying out subtraction on wavelet coefficient matrixes of corresponding directional son-bands of the filtered images with the same decomposition level number; carrying out enhancement and two-dimensional wavelet inverse transformation reconstruction on wavelet coefficient difference matrixes in horizontal and vertical directions by using a sobel operator; and fusing the reconstruction images with different decomposition level numbers so as to obtain a final difference map by using a treelet algorithm, then carrying out level set segmentation on the differencemap so as to obtain a change detection result. By using the method disclosed by the invention, the accuracy of the change detection result can be improved effectively, and the edge feature of a change area can be maintained better, therefore, the method can be applied to the fields of natural disaster analysis, land resource monitoring, and the like.

A method for suppressing speckle noise in a SAR image combined with visible light image information, the steps of which are: (1) classify pixel point area attributes, and use optical image information to divide pixel points into three types: uniform area, slowly changing area and edge detail area; (2) According to the category of the pixels, the speckle noise of the SAR image classification is suppressed respectively. The present invention uses visible light images to divide SAR images into three types of image areas: uniform area, slowly changing area and edge detail area, and then uses different noise suppression operations in different image areas; Visible light image information can accurately classify the region; (2) the most suitable speckle reduction operation is used in the uniform area and the slowly changing area, which can obtain the best speckle reduction effect; (3) the edge detail area adopts edge-keeping processing, so The best edge-preserving performance can be obtained.

The embodiment of the invention discloses a computer device for removing multiplicative noise of a color image and equipment. The computer device comprises a noise removal assembly used for carrying out multiplicative noise removal on an input color image to be processed, and the noise removal assembly comprises an image collector used for obtaining the color image to be processed and a model construction module used for constructing a denoising model in advance and determining various parameters of the denoising model. The model construction module comprises a similarity processing sub-moduleused for determining the similarity between the to-be-processed color image and the denoised color image based on data items of a universal multiplicative noise model, and a smoothing processing sub-module used for determining the smoothness of the denoised color image based on rule items of a coupling MTV model and a high-order TC model; and the data processing sub-module is used for calculatingvarious parameters of the denoising model by utilizing an augmented Lagrange method and an alternate optimization multiplier method. The non-edge area of the denoised image is smooth and natural, thestep effect is eliminated, and the edge and corner points of the color image are well kept.

The invention provides a multi-focusing-image fusion method based on a bilateral filter. The method comprises the steps of firstly, decomposing a source image for obtaining a basic layer and an edge layer; then, introducing a bilateral filtering algorithm at a fusion strategy aspect for respectively determining fusion weights of the basic layer and the edge layer, thereby sufficiently utilizing spatial continuity between the basic layer and the edge layer; and finally adding the fused basic layer and the edge layer for obtaining a final fused image. The multi-focusing-image fusion method provided by the invention has advantages of reserving image information, realizing relatively high edge keeping effect, and obtaining a good subjective effect and a relatively good objective index.

The present application relates to an underwater image processing system and method with edge preservation. The system includes: a transmission map estimation module configured to perform a transmission map recovery operation on the acquired degraded underwater image to generate a transmission map estimate of the underwater image; an ambient light estimation module configured to perform a transmission map estimate on the underwater image image performing an ambient light estimation operation to generate an ambient light estimate of the underwater image; a defogging processing module configured to perform defogging processing on the underwater image based on the transmission map estimate and the ambient light estimate to obtain a defogged underwater image; and a discriminator module configured to discriminate the transmission map estimate and the defogged underwater image compared to the true value of the transmission map and the true value of the defogged underwater image, and based on the discriminant result to train and supervise transmission map estimation and ambient light estimation.

The invention belongs to the field of infrared image processing, and mainly relates to an infrared weak and small target detection method based on time-space domain background suppression. The infrared weak and small target detection method is used for achieving the aim of infrared movement weak and small target detection in a complicated background and includes the steps that firstly, stable background noise waves in a space domain are suppressed through guiding filtering; secondly, slowly-changed backgrounds in a time domain are suppressed with a gradient weight filtering method on the time domain through target movement information in an infrared image sequence; thirdly, the time domain background suppression result and the space domain background suppression result are fused to obtain a background-suppressed weak and small target image; finally, the image is split through a self-adaptation threshold value, and a weak and small target is detected. By means of the infrared weak and small target detection method, during target detection, space grey information of the infrared weak and small target is used, time domain movement information of the target is further sufficiently used, the background noise waves are suppressed in the time domain and the space domain, and therefore the movement weak and small target detection performance in the complex background is greatly improved.

The invention discloses a multi-image encryption method, a multi-image decryption method and a computer readable storage medium. The multi-image encryption method comprises the steps of placing multiple scrambled to-be-encrypted images on cross sections at different depths of a virtual three-dimensional object; carrying out Fresnel diffraction calculation on object light waves of each cross section to obtain Fresnel diffraction light waves of each cross section on a first observation plane; calculating to obtain object light waves of the first observation plane; carrying out Fresnel diffraction calculation on the object light waves of the first observation plane to obtain the object light waves of a second observation plane; and carrying out inverse Fourier transform on the object light waves of the second observation plane after the object light waves are subjected to conjugate symmetrical arrangement so as to obtain a real-value encrypted image. The decryption method comprises the decryption steps opposite to the encryption steps, and a decrypted image is reconstructed on a reconstruction image surface in non-interfering, non-interpolation and controllable manners. The multi-image encryption method can realize multi-image encryption and has relatively high safety and robustness; and the reconstructed image of the decrypted image is relatively high in quality, and has a good edge preserving characteristic.