218 results about "Salt-and-pepper noise" patented technology

The invention discloses a color digital image enhancing and denoising method under random illumination. The method is characterized by comprising the following steps of: 1, image denoising processing, namely converting an original image from an RGB space to a YCbCr space, removing a Gaussian noise by using a Gaussian filter and removing a salt and pepper noise by performing median filter; 2, image brightness/contract stretch processing, converting the image from the RGB space to an HSI space, decomposing the image by using a two-sided filter, processing by an improved Retinex model algorithm, and obtaining a new image saturation by performing saturation compensation; and 3, performing fusion display on the image acquired after denoising the YCbCr space and the image acquired by HSI space processing. By the method, the color constancy of the image can be kept, the dynamic range of the image can be improved well, and simultaneously noises of the image can be inhibited and removed with less texture and detail information loss.

The invention relates to the field of image processing, and provides a novel self-adaptive secondary noise point detection method based on direction information. The method can be used for effectively reducing the erroneous judgment probability of non-noise points and removing impulse noise in images more effectively, and the denoising robustness on noise with different strengths is stronger. To this end, the technical scheme adopted in the invention is as follows: an image denoising method based on secondary noise point detection comprises the following steps: at first, determining an impulse noise density condition, counting all gray values in an image, calculating the number of pixel points whose gray values are 0 or 255, dividing the number by the number of total pixel points to obtain an approximate noise probability density pa, when pa is larger than or equal to s, filtering all points whose gray values are 0 or 255, when pa is smaller than s, carrying out secondary noise point detection on these points, and the detection principle is mainly based on direction features of edge information in the image; and then self-adaptive filtering is carried out. The image denoising method based on secondary noise point detection provided by the invention is mainly applied to image processing occasions.

The invention discloses a method of quickly eliminating composite noise in images, which can eliminate salt and pepper noise and Gaussian noise at the same time, improve the peak signal to noise ratio(PSNR) of the images and obtain more clear images. The main technical idea comprises the following steps: the median of the images containing noise is filtered to obtain the images the salt and pepper noise of which is eliminated and space self-adaption two-sided filter is carried out on the images the salt and pepper noise of which is eliminated to obtain the images the Gaussian noise of which is eliminated, i.e. the final noise-free images. The calculation, data processing of the algorithm of the invention is definitely adapted to the real-time image quick process. The technical flow can be adjusted according to concrete application and requirement.

The invention provides a method for extracting finger vein features for matching recognition, which comprises the following steps: (1) collecting finger vein images with an image acquisition device; (2) pre-treating the collected finger vein images, comprising making color images gray, extracting finger areas, eliminating salt-and-pepper noise and gaussian noise respectively with a combined filter, splitting images and performing binaryzation for split images through a local dynamic threshold algorithm, de-noising in an area elimination method, extracting finger vein context images according to finger contour markers, and finally standardizing the sizes of the images to form unified images; (3) extracting finger vein features through two-dimensional linear discriminant analysis algorithms weighted both in the line direction and in the row direction; and (4) carrying out matching and recognition with a nearest classifier. The invention significantly improves the speed of finger vein recognition and achieves stable and high recognition rate.

The invention discloses an image impulse noise removing method for preventing edge blurring. Edge pixel determination, impulse noise determination and filter processing are carried out respectively; an edge pixel is processed independently; and noise points in other pixels are processed by means of consistent weight mean filtering by selecting filter windows with different sizes. According to the method, edge pixel determination is carried out; an edge pixel is determined according to the characteristic of a large difference of pixel values around the edge pixel point of the image; noise removing is carried out by using an independent way, so that image edge information can be kept well to prevent an edge blurring phenomenon. During an impulse noise filtering process, no fuzzy rule needs to be designed; and because no threshold selection needs to be carried out, the calculation efficiency is improved. The method is suitable for noise removing of images with different pollution degrees.

The invention discloses an image processing-based bottled liquid impurity detection system and a detection method thereof. The detection system consists of a mechanical part and a machine vision part; the mechanical part comprises a step motor, a conveyer belt, a fixed pulley and a mechanical clamping device; the machine vision part comprises a camera obscura, a related internal structure, a FPGA (Field Programmable Gate Array) board, a serial communication port, a display module, an alarm module and upper computer software. The detection method of the system is as follows: an image is acquired; the image is segmented from the background by the seed fill algorithm; salt and pepper noises and interference pixels are filtered by fuzzy processing; contrast is increased by binarization processing; the image is connected completely by erosion and dilation operation to remove tiny noise, so that impurities become visible; finally, contour tracing and extraction are carried out to obtain the number, shapes and positions of the impurities. The invention increases image acquisition and processing speed, consequently, manual detection can be replaced, and the purpose of increasing detection speed, quality and production benefit is achieved.

The invention relates to an image processing method for filtering mixed noise and belongs to the image processing technical field. The image processing method for filtering the mixed noise comprises enabling a filter window center and pixel points of coordinates in the image to achieve superposition by building the filter window; obtaining mid-values of gray values of all the image pixel points in the filter window and determining a range of the gray values by utilizing threshold values; obtaining mean values as pixel point gray values of the image coordinates according to the sum of the pixel gray values and the pixel point numbers within the obtained range of the gray values; obtaining all the pixel point gray values of all the coordinates of the entire image to achieve the entire image processing in the same manner. The image processing method for filtering the mixed noise has the advantages of being capable of effectively filtering the impulse noise by choosing the mid-values as the center of the gray level window; restraining the gaussian noise by setting the threshold values to determine the range of the gray values according to the actual conditions and performing average on the pixel point gray values which fall within the range; not only effectively filtering the mixed salt and pepper noise and the gaussian noise, but also being simple in algorithm and being small in calculated amount simultaneously and being beneficial to the real-time application of the engineering.

The invention discloses an improved Canny image edge detection method with noise immunity. On the basis of a Canny algorithm, the improved algorithm combines adaptive median filter to replace Gaussianfilter to denoise the image, so as to filter salt and pepper noise interference better. Then combined with the maximum inter-class variance method and the maximum entropy method, the double thresholdselection method is improved to get the high and low thresholds, so as to achieve the edge detection of the image, so that the target image retains the edge information as much as possible, while filtering out the unnecessary interference edges. The invention realizes the edge detection of an image under the condition of salt and pepper noise pollution and a double threshold selection method withstrong adaptability, and has high application reference value.

The invention discloses a method for removing salt and pepper noise. The method comprises the following steps: carrying median filtering on an input noise image; differentiating a median filtering result and the input noise image and determining noise candidate points based on the differentiating result; carrying out edge detection on the median filtering result to obtain edge minutiae; eliminating the edge minutiae with differential absolute values less than a first threshold value in the noise candidate points, thus obtaining a real noise point; and carrying out local median filtering on the real noise point, thus forming a de-noised image subjected to salt and pepper noise removal. Correspondingly, the invention also provides a device for removing the salt and pepper noise. The device comprises a noise candidate point acquiring module, an edge minutia acquiring module, a real noise point acquiring module and a local median filtering module.

The invention discloses a density-adaptive image salt-pepper noise switching filtering method, which comprises the following steps of: training a plurality of neural networks adaptable to the accurate detection of noise positions under certain noise densities by using training images with different noise densities; rapidly and roughly evaluating the salt-pepper noise densities of the images; selecting corresponding neural networks according to the evaluated salt-pepper noise densities to perform accurate noise detection on the images; and performing window-adaptive switch weighted mean filtering on the images according to the accurately detected noise positions. By the method, the magnitudes of the noise densities can be automatically judged, and the accurate noise detection can be performed by adopting the corresponding neural networks for the densities with different magnitudes.

The invention discloses a two-dimensional Renyi entropic threshold segmentation method for grayscale images, which comprises the following steps: carrying out median filtering on an input original image I so as to obtain an image I_med subjected to median filtering; based on the original image I and the image I_med subjected to median filtering, constructing a grayscale median two-dimensional histogram; carrying out optimization on the obtained two-dimensional histogram by using an inertia weight increasing strategy introduced particle swarm method so as to obtain the best threshold and the optimal segmentation point; and according to the best threshold and the optimal segmentation point, carrying out segmentation on the image. According to the invention, through introducing a median filtering technology with good robustness in image filtering, replacing a window grayscale mean in the traditional method with a neighborhood window grayscale median, and combining with the original image, a novel two-dimensional histogram is constructed; an objective function is optimized by using the inertia weight increasing strategy introduced particle swarm method, so that the effective rapid segmentation is performed under noisy conditions, thereby accelerating the convergence speed and improving the segmentation efficiency.

The invention discloses an unsteady measurement algorithm based image segmentation method of an improved rule distance level set, belongs to the field of digital image processing and aims to more precisely segment images of gaussian noise and salt and pepper noise interference. The method includes: firstly, constructing a mean operator, and setting a stopping function by the constructed operator; secondly, manually setting an initial contour, and initializing a level set function according to the contour; and thirdly, importing the stopping function set in the first step into an energy equation of a DRLSE model, minimizing the energy equation by the aid of a central difference method, and iterating by taking the initialized level set obtained in the second step as an initial condition to obtain a zero level set of a steady state solution, namely the final segmentation results. The method is more precise in segmentation of the images of the gaussian noise and salt and pepper noise interference as compared with a traditional geometric Snake model method.

The invention discloses a visual guidance welding robot weld joint fast recognition technology based on double dynamic windows. According to the visual guidance welding robot weld joint fast recognition technology, two layers of dynamic windows are used for selecting an image processing area, image data processing capacity is effectively reduced, and the technology is suitable for occasions with high real-time requirements. According to the similitude of sequence images, adjacent sequence images are subtracted from the first layer of dynamic window to conduct multiplication, and large random interference is effectively eliminated; a median filtering method is adopted to filter impulse interference and impulse noise; binaryzation is conducted through the Otsu method, and processing is simple and rapid; Hough transformation is adopted to conduct line extraction, the precision is high, and interface resisting capability is good; the gravity center of a characteristic triangle formed by characteristic angular points is used as a weld joint characteristic point, and solution is simple and rapid. The visual guidance welding robot weld joint fast recognition technology can better meet the requirements of an industrial scene for accurate and rapid extraction of the weld joint characteristic point.

The invention relates to a filtering noise reduction system and a filtering noise reduction method based on an FPGA (field programmable gate array) platform, wherein the filtering noise reduction system comprises a template generating module, a data analysis module, a median filtering module, a mean filtering module and a data output module, wherein the template generating module comprises four paths of FIFO (first in first out) memories FIFO1 to FIFO4 and a path of data input, continuous four-frame video image data led by the data input are stored in the four paths of FIFO memories FIFO1 to FIFO4, and a 5*5 neighborhood template is formed; the data analysis module judges the noise type, data in the 5*5 neighborhood template is selectively transmitted to the corresponding median filtering module or the mean filtering module according to the noise type, salt and pepper noise is inhibited and removed through filtering by the median filtering module, or Gaussian noise is inhibited and removed through filtering by the mean filtering module. The system and the method have the advantages that corresponding filtering modes can be automatically selected according to different types of noise in videos, and the real-time noise reduction processing on the videos is carried out.

The invention discloses an image de-noising method by combining bidimensional Hilbert transformation with BEMD (bidimensional empirical mode decomposition). The image de-noising method includes steps of performing BEMD for an image containing noise to obtain an IMF (intrinsic mode function) and R (residue); enabling IMF components to be subjected to the bidimensional Hilbert transformation; and reconstructing images and residues which are obtained after the Hilbert transformation to obtain a de-noise image. The image de-noising method has the advantages that the image can be decomposed by means of BEMD in a multi-scale manner, a local time-frequency characteristic is realized, problems caused by one-scale de-noising by the traditional filter method are solved, and the image decomposition efficiency and efficiency are improved; and the bidimensional Hilbert transformation can realize excellent filter effects for multiplicative noise, Gaussian noise and salt-and-pepper noise of images, influence of the noise in the images to details and critical contents of the images is weakened, and the noise can be effectively removed.

The invention discloses a switching non-local total variation based filtering method for an image polluted by salt and pepper noise, which comprises the following steps: performing pre-processing for the image polluted by salt and pepper noise through a two-stage morphological detector, so as to obtain a reference image and noise flag bit; based on the noise flag bit, filtering the reference image by utilizing an improved non-local total variation method, so as to obtain a denoised image. Due to the adoption of the method, noise distribution condition of the image can be effectively detected, with very lower omission ratio and false detection ratio. Besides, based on the reference image, the similarity between two images can be accurately calculated, the salt and pepper noise can be effectively suppressed while specific information including the image edge and texture can be well protected, and optimized peak value signal-to-noise ratio and structure similarity can be offered compared with a conventional filtering method for the salt and pepper noise.

The invention relates to a method for removing salt-pepper noises in images, which comprises the following steps: A, identifying whether pi, j is a salt-pepper noise point; B, when the pi, j is a salt-pepper noise point, acquiring a 3*3 local window omega 3 centering on the pixel pi, j, and estimating a noise density d (omega 3) of the omega 3; C, when the noise density d (omega 3) obtained in the step B is less than 9, defining a filter output ri, j of the pi, j as follows: omega (fs, t, fi, j) = exp (-beta |fs, t - fi, j |); and D, when the noise density d (omega 3) obtained in the step B is equal to 9, defining the filter output ri, j of the pi, j as the steps in the specification. The invention aims to provide a method for removing salt-pepper noises in images, through which salt-pepper noise points in an area adjacent to some pixel can be removed better, so that adverse effects on the filter output of center pixels caused by noise points in local windows us reduced, and a more accurate filter output is obtained by using a new direction weighted average strategy.

The invention realizes fruit sorting on two aspects such as hardware design and model selection and software system development by using a machine vision technology, and comprises the steps of designing and selecting a machine vision industrial camera, a lens and a lighting scheme to ensure the fruit sorting system to continuously and steadily obtain a high-quality fruit image; suppressing salt-and-pepper noise and Gaussian noise in the fruit image by using median filtering and binomial filtering; highlighting edge information of the fruit image by adopting an anti-sharpening mask method at the same time; realizing fruit classification by adopting a supervised Gaussian mixture model clustering algorithm; and calibrating data grasped by a robot guided by a monocular camera through a Zhang Zhengyou calibration method, and calculating the actual grasping position and pose.

The invention relates to a low-false-alarm-rate color QR code digital watermarking method, which is a quick and efficient double color image watermarking algorithm based on color QR codes and DWT-SVD technology. Tone-level-rich color QR codes are normalized and embedded to a carrier image obtained after DWT conversion and SVD decomposition through different channels, so that the watermarking method not only has good invisibility, but also can be resistant to conventional attacks of cutting, rotation, JPEG compression, Gaussian noise and salt-and-pepper noise and the like, and has good robustness and higher QR code decoding rate. Insertion of a lot of information can be realized through insertion and extraction of the tone-level-rich color QR codes, thereby meeting requirements of high-end companies for the watermarking algorithm; meanwhile, the method also has faster operation speed, so that efficiency is improved greatly; and through insertion of a check mark, under the condition that the watermarked carrier image obtained after attacks can be extracted normally, the problem of high false alarm rate of an SVD decomposition algorithm is effectively solved. The watermarking method has a certain application value in digital right management.

The invention discloses an unmanned aerial vehicle remote sensing image segmentation method based on color space classification. According to the method, a high-resolution unmanned aerial vehicle remote sensing image is divided into a series of pixel clusters. After that, pixel points are classified based on the similarity of color space coordinates and texture features. Finally, the fusion of pixel clusters is guided in combination with the classification result. As a result, it is guaranteed that each partition basically belongs to a homogeneous region. The problems of low classification precision, large pepper salt noise and low algorithm efficiency existing in the prior art are solved. The purpose of quickly and accurately segmenting the high-resolution unmanned aerial vehicle remote sensing image is realized.

The invention discloses a retinal fundus image preprocessing method. The retinal fundus image preprocessing method is characterized by including the steps of 1), reading an original image, namely reading the original image of the retinal fundus through a green channel; 2), removing blood vessel center light reflex; 3), removing salt-and-pepper noise; 4), smoothing noise; 5), conducting background extraction; 6), obtaining a shadow correction image; 7), obtaining a homogenization image; 8), solving a complementary image; 9), obtaining a blood vessel enhanced image; 10), outputting the enhanced image. The retinal fundus image preprocessing method has the advantages that the method enables the processed retinal image to be better in image quality and richer in information quantity, reserved enhancement of useful information such as blood vessels, optic disks and central foveae in the retinal image is achieved, and feature extraction, segmentation and recognition of the retinal image are improved.

The invention provides a remote sensing image change detection method based on significance detection and a deep twin neural network. The remote sensing image change detection method is characterizedby comprising the following steps: preprocessing a two-time-phase remote sensing image; carrying out normalization processing on the difference image; carrying out multi-scale segmentation and mergingoptimization; obtaining a saliency detection graph; establishing a double-window deep twin convolutional network model and training the double-window deep twin convolutional network model; and fusingthe segmentation object and the pixel-level change detection result through judgment to finally obtain a change detection result graph. Deep learning is successfully applied to the field of change detection, salt and pepper noise existing in high-resolution image change detection is reduced, and the precision of change type detection is effectively improved.

The X-ray imaging system of this invention includes: a detecting member which detects a salt and pepper noise region in the reconstructed image based on at least one characteristic value of the moire stripe image not including the object and/or the moire stripe image including the object; a masked-image generating member which generates a masked image for identifying the detected salt and pepper noise region; and an image processing member which masks or trims at least one of the reconstructed image and the moire stripe images with the generated masked image.