253results about How to "Clear details" patented technology

A system and method for road surface defects detection based on infrared imaging technology, while the system comprising: a detection vehicle traveling on a detected road surface; a pan and tilt provided on the detection vehicle and rotate horizontally and vertically with respect to the detection vehicle; an infrared camera detachably set on the pan and tilt, which is used to capture infrared thermal images of the detected road surface, and to output inferred thermal image digital signals about the inferred thermal images including temperature values of the detected road surface; a main controller provided on the detection vehicle and connected to the pan and tilt and the infrared camera respectively, which is used to control capture actions of the infrared camera, to control angles of the horizontal and vertical rotations of the pan and tilt, and to transform the infrared thermal image digital signals output from the infrared camera into digital signals to be used in standard network transmission; and a data processor, used to receive the digital infrared signals to be analyzed and processed to determine the type and location of defects on the detected road surface. The system of the present application has excellent operation flexibility and is able to visually display details of the defects.

The present invention relates to a method capable of real-time implementable local tone mapping of high dynamic range images. In one embodiment, the present invention relates to a method capable of accomplishing real-time local tone mapping of high dynamic range images so that they have clear details without, for example, halo artifacts when viewed on standard display devices. In another embodiment, the present invention relates to a method capable of accomplishing real-time local tone mapping of high dynamic range images so that they have clear details without, for example, halo artifacts when viewed on standard display devices where such a method utilizes, in part, a modified Reinhard operator.

The invention discloses an image super-resolution reconstruction method for generating antagonistic network based on feature fusion, which comprises the following steps: pre-processing ImageNet data set to obtain a reconstruction data set corresponding to high-resolution image and low-resolution image; The model of generating antagonistic network for training is constructed, and the interpolationreconstruction module is introduced into the model to construct the generating network and perceptual network for multi-feature fusion. The reconstructed data sets are sequentially input into the generated antagonistic network for model training. The images to be processed are normalized to obtain low-resolution images, which are input to the trained generation network to obtain the reconstructedhigh-resolution images. As that recursive residual network is utilize to extract multiple features such as edge and texture, highlight the edge and texture information of the generated image, so thatthat reconstruct image is clearer, The interpolation reconstruction module is used as a mediator to realize the residual discrimination, highlight the difference between the reconstructed and real high-resolution images and reduce the computational complexity of the discrimination network, and use VGG network to fuse multiple features to calculate the loss function, which makes the reconstructed image more effective and more consistent with the observation mode of human eyes.

The invention relates to a microscopic image fusing method based on two-dimensional empirical mode decomposition, which comprises the following steps: performing multi-scale decomposition on the acquired ordered microscopic original images by using a two-dimensional empirical mode decomposition method, thereby acquiring the multi-scale high-frequency components of original images; fusing the multi-scale high-frequency components according to local obvious standard; fusing the low-frequency components of the original images by using a principal component analysis method; and finally reversely recomposing to acquire a merged image. By using the method provided by the invention, the multi-scale decomposition is performed on the acquired ordered microscopic original images by using the two-dimensional empirical mode decomposition method and the decomposition process is adaptive; high-frequency fusing treatment is performed according to the local obvious standard based on a big area value and the relevance between adjacent coefficients is considered, so the detail information clearly focused of each original image can be supplied; and the low-frequency fusing treatment is performed by using the principal component analysis method, so the relevant information of original image pixel is utilized and the visual decoding effect of the merged image is increased, thereby increasing the quality of the fused image.

The invention discloses a morphology and PCA (principal component analysis) based contourlet fusion method for infrared and visible light images. The method includes: firstly, subjecting two images to morphology-hat transformation respectively and subjecting the two obtained images to contourlet decomposition to form a high-frequency image and a low-frequency image respectively; then, fusing the low-frequency image by a PCA method; fusing the high-frequency image by means of average gradient adaptive weighted processing; reconstructing a fused image by means of contourlet reverse transformation. The morphology and PCA based contourlet fusion method for the infrared and visible light images is used for image fusion of the infrared and visible light images, so that the fused image is high in contrast ratio, rich in information and clear in texture detail.

The invention provides an SAR image synthesis method for enhancing target characteristics. According to the method, an SAR image and an optical image are effectively utilized for conducting synthesis, and therefore the target characteristics are enhanced. According to the technical scheme, the method comprises the steps of firstly obtaining the optical image and the SAR image shot from the same scene, conducting color preprocessing on the optical image through a color conversion module in a computer, utilizing a constant false-alarm for detecting high bright spots in the SAR image, conducting region merging on the bright spots, then, utilizing target prior information and the characteristics for completing detection post-processing, removing the false alarm, dividing the SAR image into a target region and a background region, utilizing the SAR image target position information by an optical image target characteristic extracting module for extracting a target region and the target characteristics of the optical image, finally, conducting three-channel color image synthesis on the SAR image and the optical image through an image synthesis module, and overlaying the extracted target edge of the optical image to the synthesis image for generating a final synthesis image.

The invention relates to a display screen display optimization method. The method comprises the following steps that: (a) current ambient light color temperature and current environment light brightness of a display screen are acquired in real time; (b) the color temperature adjustment target value of the display screen is calculated through utilizing a binary linear equation based on the current ambient light color temperature; (c) the brightness adjustment target value of the display screen is calculated through utilizing a piecewise function based on the current environment light brightness; and (d) the color temperature and the brightness of the display screen are adjusted according to the color temperature adjustment target value and the brightness adjustment target value. With the method adopted, parameters such as color temperature and / or contrast of the display screen can be controlled when the brightness of the display screen is adjusted, and therefore, the display of the display screen can better accord with the change of surrounding environment of the display screen, and image display quality can be improved.

An embodiment of the invention provides a backlight module and a display device and relates to the technical field of display with an aim to realize repeat adjustment of LED lamps during regional dimming. The backlight module comprises a light source assembly and a light-rendering film arranged on the light emerging side of the light source assembly; the light-rendering film comprises a plurality of dimming unit arrayed in matrix form; the dimming units are used for receiving dimming signals and changing transmittance of the dimming units according to the diming signals. The backlight module is used for providing backlight.

The invention discloses a pixel circuit and a driving method thereof, and a display panel. The pixel circuit comprises a first sub-pixel circuit and at least one second sub-pixel circuit, wherein the first sub-pixel circuit writes a first data voltage provided by a first data line under control of a first scanning line and generates a compensation voltage at a first node; the at least one second sub-pixel circuit carries out threshold voltage compensation through using the compensation voltage generated at the first node; and based on a display mode, the at least one second sub-pixel circuit writes a second data voltage provided by a second data line under control of a second scanning line. By using the pixel circuit and the driving method thereof, and the display panel, combination of threshold voltage compensation and intelligent display is performed on a driving transistor of the pixel circuit so that an influence of drifting of a threshold voltage on a driving current of the driving transistors can be eliminated, and visual resolution to each area of the display panel is adjusted in real time aiming at a condition that a user pays attention to different points of a frame displayed by the display panel.

The invention discloses an underwater polarization image fusion system based on a bionic vision mechanism. The system comprises a cluster light underwater polarization imaging module based on a non-uniform light field, a polarization parameter image calculation module, an image storage module, an image grading fusion module and an output display module. First of all, the non-uniform light field can be established by taking a cluster light source as underground illumination, underwater polarization images are extracted by use of a bionic mantis and shrimp vision polarization perceiving mechanism, and thus backward scattering caused by particles suspended in water is inhibited; and then, by use of information correlation and complementarity between polarization parameter images, based on a human eye vision feature, a grading fusion model of the images is constructed, and thus feature and multi-scale fusion of the polarization images is realized. According to the invention, the sharpness and the contrast of the underwater polarization images can be effectively improved, detection and analysis of underwater targets are facilitated, and the detection precision of the underwater targets is further improved.

The embodiment of the invention discloses a mammary gland image processing method and a mammary gland image processing device. The method comprises the following steps: extracting a gland area of the mammary gland image; obtaining the gland distribution characteristics of the gland area; comparing the gland distribution characteristics with the preset mammary gland classification standard, so as to determine the image type of the mammary gland image; determining the preset mammary gland image processing manner corresponding to the image type, and utilizing the mammary gland image processing manner to process the mammary gland image. Compared with the prior art, the mammary gland image processing method is to distinguish and process different types of the mammary gland images according to the gland distribution characteristics of the mammary gland image, so that the processing result is more targeted, and the accuracy rate is higher.

The invention provides a PET image reconstructing method. The method comprises the steps that PET projection data Y are acquired; an initial estimation image is acquired according to the PET projection data Y; iteration reconstruction is carried out on the initial estimation image to obtain a target estimation image; a target function is established based on total variation sparse conversion and a prior image Xp according to the target estimation image, and the expression of the target function is shown in the specification, wherein the prior image Xp is an image obtained after iteration reconstruction is carried out through the PET projection data Y; the minimum solution of the target function is solved under the constraint condition of A*X=Y, and the minimum solution of the target function is the reconstructed PET target image. By means of the method, the obtained image has clear detail information and high image contrast, and the wide application range of the PICCS algorithm is enlarged in the field of PET image reconstruction. The invention further provides a PET image reconstructing device.

The invention provides an image display control method and a terminal, and relates to the technical field of terminals. The image display control method is applied to a terminal and comprises: receiving a first input of a user to a first image, wherein the first input is used for zooming the first image; in response to the first input, updating the first image into a second image for display, wherein the first image is an image acquired by a camera under the condition of a first focal length, the second image is an image acquired by the camera under the condition of a second focal length, andthe first focal length is not equal to the second focal length. According to the scheme, when a user zooms the image, the image can be replaced by another image with a clearer visual angle, and details can be ensured to be clear when the image is zoomed and viewed.

The invention provides a multifunctional image processing method based on wavelet transform. The multifunctional image processing method comprises the following steps: step 1, reading an original image; 2, decomposing the original image into a high-frequency part and a low-frequency part by wavelet transform; 3, for the high-frequency part of the image, performing threshold quantization processingon all high-frequency coefficients, and then performing median filtering to complete compression of the high-frequency part and image enhancement; 4, for the low-frequency part of the image, enhancing a low-frequency coefficient by adopting an improved function; and step 5, reconstructing the processed high-frequency part and the processed low-frequency part by using wavelet inverse transformation to obtain a reconstructed image. According to the method, wavelet transformation is adopted to process the image, so that the entropy after signal transformation is reduced, the non-stationarity ofthe signal can be well described, and feature extraction and protection are facilitated. According to the method, wavelet transform is adopted, so that denoising is more facilitated in a wavelet domain than in a time domain, and different wavelet functions can be selected according to different application requirements to obtain an optimal processing effect.

The invention discloses a method for eliminating cloud and haze of an atmospheric degraded image. The method comprises the steps of the acquisition of a dark channel image and mid-value filtering, the self-adaptive degradation and acquisition of an atmospheric light image, the detailing of an atmospheric transmission function and the visual correction of a color domain. The atmospheric degraded image to be processed by the method has the problems of different illumination intensities, contrast ratios and dynamic ranges, a dark channel theory model is adopted, and the mid-value filtering of the dark channel image, the self-adaptive degradation and acquisition of the atmospheric light image and the visual correction of the color domain are also combined to eliminate the cloud and haze, and the disadvantages in the prior art are overcome. Besides, the method for eliminating the cloud and haze of the atmospheric degraded image is simple to operate and has good application prospect.

The invention discloses a low-illuminance image classification method based on an attention mechanism and a capsule network, and solves the technical problems that a low-illuminance image shot under weak light is too dark, so that the visual effect and the image classification result effect are poor, and detail loss and noise influence the classification of the image. According to the method, a CBAM attention mechanism network is utilized to enhance the brightness of a low-illumination image, remove the noise of the image and improve the details of the low-illumination image, and then the enhanced image is input into a capsule network for classification. According to the method, the brightness and noise of the low-illumination image are optimized through the attention mechanism and the capsule network, the image with enhanced brightness and clear details is obtained, and the image effect in the classified weak light environment is better.

The invention discloses a scene analysis method based on binocular vision, and the method comprises the steps: obtaining a first original image and a second original image corresponding to the binocular vision, and extracting the multi-scale features of the first original image and the multi-scale features of the second original image; combining the multi-scale feature of the first original imagewith the multi-scale feature of the second original image, performing depth estimation operation, and outputting primary depth estimation information of the first original image based on the combinedmulti-scale feature; performing semantic segmentation calculation by using the multi-scale features of the first original image, and outputting primary semantic segmentation information correspondingto the first original image; and fusing the primary depth estimation information, the primary semantic segmentation information and the multi-scale features of the first original image, and outputtingfinal depth estimation information and semantic segmentation information. The calculation efficiency can be improved, the calculation speed is increased, and the calculation precision is improved.

The invention discloses a high-speed railway foreign matter identification and classification method based on a GMM model and a BP neural network. The high-speed railway foreign matter identification and classification method comprises the steps: obtaining a real-time video image of a target object through a high-definition camera, observing the final state and position of the real-time video image, and judging whether the target object is a foreign matter or not; and identifying and classifying foreign matters based on a Gaussian mixture model and a BP neural network, so that the types of different foreign matters are obtained and warning information is sent to a train. According to the invention, classification of invaded foreign matters can be realized, and recognition of the foreign matters is more accurate, and early warning can be carried out in time, and the accuracy of foreign matter recognition is improved to a great extent, and more detailed information is provided for subsequent removal of the foreign matters.

The invention belongs to the technical field of oilfield seismic big data reconstruction, and particularly relates to a seismic data reconstruction method based on spatial constrained compressed sensing. The method comprises the steps that a part of data is taken as training data, K-SVD dictionary learning is used for training an overcomplete dictionary to reconstruct original earthquake data; a joint sparse decomposition method is used for extracting shared spatial information, and a sensing matrix in a compression sensing algorithm is transformed; a sparsity self-adaptive matching tracking algorithm is improved, an initial sparsity estimation method is introduced, and the data is reconstructed by adopting a variable step size strategy. The reconstructed result not only is clear in detail, but also greatly reduces the operation time compared with IRLS and SAMP, and the horizontal transition is smoother, which indicate that the designed algorithm utilizes spatial related information and the reconstructed result is more real.

The invention discloses a flow cytometer based on a micro-fluidic three-dimensional focusing technology. The flow cytometer comprises a micro-fluidic chip; a parallel light source is arranged over themicro-fluidic chip, and a lensless sensor is arranged on the lower surface of the micro-fluidic chip; and the lensless sensor is connected with a data processing module. Three-dimensional focusing ofcells is achieved, and thus a cell image is convenient to collect; and in a microchannel of a main fluidic chip, because of blocking of a triangular obstructing tiled plate, flow clamping focusing liquid is formed in a focusing channel, thus cell sap is focused for the first time, sample buffer solutions on the two sides of an outlet of the focusing channel flow out from a sheath fluid channel tofocus the cell sap flowing out from the focusing channel again, thus the cells in the cell sap evenly flow through a lensless sensor image collecting area in single rows, lensless imaging and algorithm processing are facilitated, a collected cell image is clear in detail and high in resolution ratio, and the problems of cell adhesion and transcending flow are avoided.

A novel composition (A) comprising a polyamide matrix contains at least 2 (preferably 2-50) wt. % electrically conductive filler (I) and at least 1 (preferably 1-30) wt. % antistatic agent (II). Independent claims are included for the preparation of (A).

The invention relates to the technical field of image processing, and provides a rapid defogging method and system for a single image. The method comprises the following step: carrying out minimum value filtering on an inputted fogged image I(x), calculating to obtain an initial dark primary color image M(x), calculating an atmospheric transmission function t(x) according to the calculated initial dark primary color image M(x), combining a projection method to calculate an atmospheric light value A, and recovering a fog-free image J(x) according to an atmospheric scattering module of the fogged image and the calculated parameters of the atmospheric light value A and atmospheric transmissivity. Therefore, the fogged image is rapidly defogged, the details of the defogging effect are clear, the color is real and natural, the method is extensively applicable, the balance between the recovery quality and the recovery speed of the image is guaranteed, and the method can be applied to an occasion operating in real time.

The invention discloses resurfacing method and device of a Windows user interface, relates to the technical field of image processing of a computer, and provides a novel resurfacing scheme of the user interface. A picture of the resurfaced user interface is exquisite; the detail of the resurfaced user interface is clearly displayed; the color of the resurfaced user interface is natural, and the user experience is improved. The resurfacing method of the Windows user interface provided by the embodiment of the invention comprises the steps of analyzing data type in a skin image of the Windows user interface to be drawn, dividing the skin image into at least two layers of layered images according to the data type, and orderly drawing the layered image according to received resurfacing command message from a user, so as to resurface the user interface.

The invention discloses a video enhancing method based on video frame self-adaption. The method comprises video frame self-adaption parameter adjustment and video frame enhancing processing. The invention also discloses a video enhancing system. The system comprises a parameter adjusting module and an enhancing processing module. According to the video frame self-adaptive video enhancing method based on an airspace point algorithm and the video enhancing system built based on the method, through starting from a performance and a video continuous frame enhancement flexibility, the performance is ensured; moreover, adjusting parameters are selected according to the video contents self-adaptively; the problem of video frame distortion resulting from excessively improving too dark images or excessively reducing too bright images by directly using conventional methods such as histogram equal saturability enhancing is solved; the enhancing of the processed video under various environments such as indoor, outdoor, low illumination and high illumination environments, is more flexible; and compared with the original video, the processed video is more clear in detail and more abundant in contrast and saturability and has better eye watching feelings.

The invention provides an image segmentation method for enhancing edge and detail information by utilizing feature fusion, and relates to the technical field of computer vision. The method comprises the steps of extracting features of an input image by using a convolutional neural network; inputting the extracted features into a decoding structure added with more feature fusion, and enriching theedge and detail information while recovering the image resolution to obtain a dense feature map; outputting maximum values of different classifications through a normalization method; and calculatinga cross entropy loss function, and updating the weight in the network by using a random gradient descent method. According to the method, the position and boundary detail information lost in the encoding stage can be recovered while the resolution of the feature map is recovered, the information of the picture is enriched, the dense feature map is obtained, the sparse feature map brought by directup-sampling is compensated, the segmented boundary and details are clearer, and the segmentation effect on detail small objects is improved.

The invention discloses an image super-resolution reconstruction method of visual vocabularies and based on texture context constraint and belongs to the field of image super-resolution research. The low-resolution visual vocabulary to the high-resolution visual vocabulary one-to-one mapping geometrical characteristic symbiotic prior knowledge is obtained through combining and aggregating to lead the similarity of geometrical characteristics to keep between each pair of low-resolution visual vocabulary and high-resolution visual vocabulary, the postulated condition based on an instance method is met, and detail accuracy of restored high-resolution images is ensure; and a two-step image super-resolution reconstruction frame is built, the frame effectively accesses the visual vocabulary which the real high-resolution images belong to through a texture context characteristic in a maximum a posteriori method, and high-resolution image details are constructed through low-resolution images by aid of subspace constraint and reconstruction constraint. Compared with the existing method, the high-resolution images extracted in the image super-resolution reconstruction method are real and clear in details and accord with the practical conditions.

The invention provides an optical projection tomography device and a method based on a camera array. A light source array and the camera array are utilized to realize two-dimensional projection images of different exposures of an organism, and high-quality reconstruction of a three-dimensional image of a biological sample is realized through a high dynamic range imaging technique and a filter back projection algorithm. According to the optical projection tomography device and the method based on the camera array, an imaging module mainly comprises a scanning acquisition module and a data processing module, wherein the scanning acquisition module mainly comprises: (1) a light source module formed by the light source array, an optical lens, a beam expander and a polarizer; (2) a biological sample platform formed by a stepping motor rotating platform and a sample clamping device; (3) a two-dimensional projection image acquisition module formed by a microscope, the camera array, a polarization analyzer and a filter plate; the data processing module mainly comprises: (1) high dynamic range image synthesis; (2) rotary center position correction; (3) filter back projection three-dimensional reconstruction.

The invention discloses a segmentation method of a high-noise gray-scale non-uniform image. The segmentation method is an image segmentation method based on combination of a distance maintenance level set method and a mark watershed method. In the segmentation method of the invention, advantages of the above two methods are used to make up mutual defects; segmentation of a target containing a high noise and a non-uniform gray scale simultaneously is realized; an edge of an interested object can be accurately segmented; and an excess segmentation problem of the image in a traditional watershed method can be overcome. Compared to a traditional level set method, by using the segmentation method of the invention, an operation speed is fast and operation time of an algorithm is not changed along with changes of an image size.

The invention discloses an underexposure image recovery method based on deep learning, and belongs to the field of image quality enhancement. The problems of low shooting contrast, low brightness, much noise, high complexity of a mainstream image recovery algorithm and high calculation cost caused by uncomfortable light source in the shooting process are solved. The method mainly comprises the following steps: firstly, the input of a multi-scale decomposition network for shooting images with different illuminance is a low-illuminance image, and a series of feature maps smaller than the original image are obtained through down-sampling convolution; secondly, the reflectivity and illuminance of images with different resolutions are obtained at an up-sampling decoding end through decomposition, on the basis of the decomposition, image brightness enhancement is carried out by using a multi-scale brightness adjustment network oriented to image illumination imbalance, and then reconstructionand recovery are carried out by using an image reconstruction network; finally, the reflectivity image and the enhanced illuminance image are integrated to obtain a final restored illumination image.