27 results about "False color" patented technology

The invention discloses a photovoltaic power station defect detection method based on thermal imaging data, and relates to the technical field of artificial intelligence, and the method comprises the steps: S1, collecting the thermal imaging data of a photovoltaic module under different working conditions, and carrying out the preprocessing of the thermal imaging data, and obtaining a thermal imaging data set; s2, based on the thermal imaging data set, dynamically dividing a temperature interval and calculating an adaptive bandwidth, and performing weight distribution and color band center weighted fusion by using a Sigmoid function to generate an adaptive pseudo-color image capable of enhancing local contrast; s3, constructing a double-branch defect detection model based on the original temperature data and the pseudo-color image; and S4, performing defect detection on the real-time thermal imaging data based on the double-branch defect detection model to obtain a defect detection result of the current photovoltaic power station. The method can be effectively applied to large-scale inspection operation and maintenance of the photovoltaic power station.

The application provides a RAW domain non-local mean image denoising method, comprising the following steps: S1, collecting a RAW image by a sensor; S2, dividing the image into N channels, such as R, Gr, Gb and B channels according to an actual Bayer format; S3, performing non-local mean filtering on each channel; further comprising: S301, calculating the similarity between each channel pixel point and the center pixel point of a neighborhood window; S302, calculating the weight according to the similarity information; S303, calculating the filtering result of each pixel point in each channel according to the weight; S4, merging the results of each channel after filtering into a RAW image, and outputting the result. The application utilizes the original characteristics of noise to perform image filtering in the RAW domain, reduces the difficulty of noise suppression, and has smaller side effects under the premise of achieving the same filtering effect; the non-local mean filtering is adopted, so that the problems of pattern noise after filtering, and false color and color cast caused by unbalanced channel filtering results are avoided.

This invention provides an image reconstruction method, an image sensor, and a terminal device. The method includes: acquiring a color filter array image collected by the image sensor; acquiring a high-frequency image and a low-frequency image based on the color filter array image; for each pixel in the high-frequency image, acquiring the region feature fusion weight and distance feature fusion weight corresponding to that pixel; for each pixel in the high-frequency image, acquiring the mixing weight corresponding to that pixel based on the region feature fusion weight and distance feature fusion weight; and fusing the high-frequency image and the low-frequency image based on the mixing weight corresponding to each pixel in the high-frequency image to generate a reconstructed image. This invention can improve the image quality noise and false colors in the edge areas while ensuring the clarity of details in the central region of the image, thereby significantly improving the overall image reconstruction quality.

The present application relates to the field of endoscope imaging, in particular to a single-channel filtering demosaicing method and system for endoscope images; the horizontal direction local gradient intensity and the vertical direction local gradient intensity of green pixels in a color filter array image are calculated, the green pixels are adaptively interpolated according to the difference between the horizontal direction local gradient intensity and the vertical direction local gradient intensity, and an initial green channel is obtained; the pixels in the initial green channel are subjected to bilateral filtering, and a filtered green channel is obtained; based on the filtered green channel, residual interpolation is used to respectively reconstruct a red channel and a blue channel, and a demosaiced image is obtained; the present application reduces the risk of false color and noise amplification, has the characteristics of strong edge preservation ability, high overall color restoration precision, simple algorithm and easy hardware implementation.

The invention discloses an image defogging method combining dark channel statistics and polarization degree constraint. The method is suitable for image enhancement and visibility improvement in a complex smog scene. According to the method, images of five wave bands in different polarization states in the same scene are collected through a polarization camera, Stokes parameters are calculated to obtain polarization degree and polarization angle information, and then atmospheric scattered light and object reflected light are distinguished. When the atmospheric light is estimated, the interference of a bright object is effectively avoided by combining the brightness and the polarization degree of a single-channel grey-scale map; in transmissivity estimation, a self-adaptive fusion strategy is provided, dark channel statistical characteristics and a polarization physical model are combined, and transmissivity estimation weight is adjusted in a self-adaptive mode according to regional polarization degree difference. And finally, realizing image restoration based on an atmospheric scattering model, and remarkably improving the information content and visual recognition effect of the defogged image by combining a multispectral false color fusion technology. Through organic combination of polarization defogging and multispectral enhancement, a complete technical scheme from image quality recovery to information depth extraction is formed, clear and information-rich image data can still be provided under severe weather conditions, and the method has important practical application value.

The application discloses a kind of sandstone core spontaneous imbibition experimental methods based on NMR scanning, comprising: using nuclear magnetic resonance scanning device scans to obtain the gray scale of sandstone core sample in initial state when not absorbing water vertically axial stratification and parallel axial direction;Using nuclear magnetic resonance scanning device scans to obtain the gray scale of sandstone core sample vertically axial stratification imaging and parallel axial direction and T2 curve under different spontaneous imbibition time under different time corresponding moment when soaking in spontaneous imbibition device;DCM data of the imaging result obtained is adjusted to window width window position, the color scale of all gray scale is adjusted to obtain the best false color map, and the hydrogen signal intensity of sandstone core sample along axial direction different position and parallel axial direction is analyzed to obtain fluid migration and occurrence state in sample space at different time in complete process of spontaneous imbibition.The application can be not destroyed core sample as premise, and space positioning is carried out by analyzing a large number of profile image data, and then movable water migration speed and reached position in spontaneous imbibition process are predicted.

Computer-aided method for creating an article-based purity definition of pre-concentrates from sorting plants for solid waste materials, the method comprising the following process steps: a) Acquisition of at least two-dimensionally resolved sensor raw data regarding at least one pre-concentrate using a sensor; b1) Classification of the sensor raw data based on predefined material classes; b2) Creation of an initial false-color image based on the classification of the sensor raw data, c) Correction of the false colors of the first false color image of article sub-components; d) Creation of a second false color image based on the change in the false color of the article component regions; e) Creation of a false color distribution based on the second false color image.

The application relates to a security X-ray image augmentation method, system, device and medium. The method estimates high-energy data and low-energy data of a contraband false color image inserted into a background false color image through a bottom-layer physical model of a security inspection machine, inserts the contraband false color image into a specified target area of the background false color image, then completes mapping of the high-energy data and the low-energy data to the false color image, generates a false color image with specific contraband inserted at a specified position, realizes TIP data enhancement based on the high-energy data and the low-energy data, can quickly generate a large amount of training data with labeled information, and achieves the purposes of directional data amplification and data enhancement.

The invention discloses a device identification and maximum temperature calculation method based on an infrared thermal imaging pseudo-color image. The method comprises the following steps: acquiring the infrared thermal imaging pseudo-color image comprising target equipment, a color gradation scale and a maximum / minimum temperature display area; identifying a target device from the image and extracting an image area of the target device; according to the pre-calibrated position, the highest temperature value, the lowest temperature value and the color gradation scale region image are extracted respectively, and the pixel height is calculated; positioning a highest-temperature pixel point in the equipment area and acquiring a color value of the highest-temperature pixel point; performing color matching on the color value and the color gradation scale image, determining the position of the color value on the color gradation, and calculating the height of a matched pixel; finally, according to the temperature range, the color gradation height and the matching height, the highest temperature of the equipment is obtained through proportional conversion, and the equipment name and the temperature result are output. According to the invention, equipment identification and highest temperature measurement and calculation can be automatically and accurately completed from a single infrared thermal image, and the automation level and accuracy of industrial temperature monitoring are improved.

The invention relates to the technical field of defect identification of image analysis, in particular to a chip packaging process defect rapid detection method and system.The method comprises the steps that hyperspectral scanning is conducted on a chip packaging area to be detected to collect a three-dimensional hyperspectral data cube, and the hyperspectral data cube after radiometric calibration is obtained through radiometric calibration processing; the method comprises the following steps: screening out a characteristic wave band combination with remarkable spectral response difference between a transparent adhesive material and a chip packaging substrate, constructing a characteristic wave band sub-data set, generating a pseudo-color image after dimension reduction mapping processing, and extracting a connected pixel set of an adhesive overflow region; calculating a physical area corresponding to the connected pixel set and a spatial position coordinate of the connected pixel set in the chip packaging area; if the physical area exceeds a preset glue overflow area threshold value, judging that the chip packaging process has a glue overflow defect and outputting a corresponding spatial position coordinate; and the detection accuracy of the transparent excessive glue defect in the chip packaging process is effectively improved.

The application discloses a forest fire situation analysis method and device based on satellite remote sensing and a medium, and belongs to the technical field of remote sensing application. The method comprises the following steps: collecting multispectral satellite remote sensing image data of a region to be researched, and preprocessing the multispectral satellite remote sensing image data to obtain a multispectral satellite orthographic image data set; in the case that a fire occurs in the region to be researched, performing false color band synthesis on the multispectral satellite orthographic image data set to obtain a false color image, and extracting a burned pixel from the false color image to determine a forest fire line contour; calculating an area in the forest fire line contour based on Arcgis software, and calculating a difference value of a vegetation normalized index before and after a disaster based on ENVI software to evaluate a fire intensity. The application realizes the identification of a forest fire through an efficient and accurate method, and timely risk evaluation after the occurrence of the fire.

The application provides an endoscope imaging method and system, an imaging method and device, and an electronic device. The system comprises: a white light illumination module for obtaining standard white light to illuminate a target tissue of an endoscope; a beam splitting module for splitting reflected light of the white light into a first light beam and a second light beam; a white light imaging processing module for obtaining a standard white light image using the first light beam; simultaneously, a hyperspectral imaging processing module for collecting a hyperspectral image using the second light beam, and selecting two or more narrowband images with a center wavelength difference reaching a preset difference from a plurality of narrowband images of the hyperspectral image; and, based on at least any two narrowband images of the two or more narrowband images, promoting and inhibiting mechanism fusion and color channel combination are performed to generate a false color image; and an image display module for displaying the standard white light image, the hyperspectral image, and the false color image, so as to realize real-time examination of the endoscope.

The invention discloses a cerebral blood flow diffusion assessment method based on attention guidance sparse image fusion, and belongs to the technical field of medical image processing, and the method comprises the steps: collecting an original data set of a CTA image sequence of an ischemic smoke disease, constructing a plurality of groups of sparse image pairs each comprising a diffusion image and a blood vessel image, and obtaining an image training set; constructing a sparse image fusion and visualization model; training a sparse image fusion and visualization model by using the image training set to obtain a trained sparse image fusion and visualization model; and processing a to-be-detected ischemic smoke disease CTA image by using the trained sparse image fusion and visualization model to obtain a corresponding reconstructed fusion image, and completing evaluation based on a pseudo-color image of the corresponding reconstructed fusion image and a visualization result. The problems that an existing method still has limitation in the aspect of sparse image fusion precision, cross-modal feature extraction and fusion efficiency is insufficient, and the accuracy of an evaluation result is limited are solved.

This invention discloses a controllable pixel displacement-driven joint demosaicing and super-resolution reconstruction method, comprising: 1. Constructing an original image sequence acquisition system consisting of a piezoelectric displacement module, a control module, and an image acquisition module, acquiring multiple frames of original image sequences through controllable sub-pixel displacement; 2. Constructing a joint demosaicing and super-resolution reconstruction network based on the multiple frame sequences, performing feature extraction and fusion on the multiple frame sequences; 3. Introducing a direction-aware wavelet enhancement module into the network, enhancing edge texture expression through wavelet decomposition, sub-band directional modulation, and reconstruction; 4. Introducing a structure-aware convolutional kernel attention module into the network, adaptively adjusting the convolutional kernel response according to local structural complexity to suppress false colors and structural distortion. This invention utilizes known displacement information to improve the determinism of original image sequence fusion and the detail fidelity and color consistency in high-frequency regions, and is applicable to scenarios such as industrial camera visual inspection and precision measurement.

The present application belongs to the technical field of coating thickness detection, and specifically relates to a kind of nondestructive testing method for uniformity of coating thickness of special anticorrosive paint, comprising: inert fluorescent tracer is incorporated into special anticorrosive paint, and linear parameters of fluorescent brightness and coating thickness are calibrated;Digital ant node is laid out, and global distributed sensing network is built. Transmit the detection signal to collect coating echo and fluorescent information, solve the coordinate and thickness characteristics to form initial data points;Global aggregated perception dataset is obtained by low-power ad hoc network transmission and convergence, and high-resolution digital twin model is constructed by edge gateway interpolation reconstruction, terahertz voxel rendering and fluorescent feature hierarchical analysis. By comparing the model thickness field with the design threshold point by point, the false color cloud picture is generated, and the coating thickness uniformity visualization detection report is output. In the present application, the layered thickness of the coating is nondestructively measured by doping fluorescent tracer and terahertz sensing, and the defects are identified and visualized modeling.

In an advanced driver-assistance system (ADAS), RAW sensor image processing for a machine vision (MV) application is important. Due to different color, e.g., red / green / blue (RGB), color components, being focused by the lens at different locations in image plane, the lateral chromatic aberration phenomenon may sometimes be observed, which causes false color around edges in the final image output, especially for high contrast edges, which can impede MV applications. Disclosed herein are low-latency, efficient, optimized designs for chromatic aberration correction (CAC) components. An in-pipeline CAC design may be used to perform on-the-fly CAC without any out-of-pipeline memory traffic; enable use of wide dynamic range (WDR) sensors; uses bicubic interpolation; support vertical and horizontal chromatic aberration color channel offsets, reduce CAC line memory requirements, and support flexible look-up table (LUT) down-sampling factors to improve the spatial precision of correction and accommodate popular image sensor resolutions.

The application discloses a kind of underwater robot image enhancement methods and systems based on sonar structure constraint generation confrontation, belong to underwater robot environment perception and multimodal information fusion technical field.The method is through underwater camera equipment and sonar sensor synchronous acquisition original underwater image data and sonar point cloud data, and pretreatment is carried out to sonar point cloud;Space guide map is constructed, and the structural boundary information and distance distribution information of different regions in scene are represented;Original underwater image and space guide map are input into generator, enhanced image is generated, and enhanced image is discriminated by discriminator, and finally enhanced underwater image is output after training;The application introduces the spatial structure information of sonar point cloud into the generation confrontation image enhancement process, so that the enhanced result can better maintain the target boundary and scene structure while improving brightness, contrast and color performance, and reduce the over-enhancement and false color problems prone to occur in traditional image enhancement methods.

The application discloses a kind of space dynamic false color display system and method based on spatial light modulator.Halogen cold light source and optical fiber light beam connection, optical fiber light beam light outlet, pinhole diaphragm, collimating lens and spatial light modulator display screen are sequentially arranged along incident light axis, spatial light modulator display screen, Fourier lens, black paper filter, Fourier lens and receiving screen are sequentially arranged along reflected light axis, and receiving screen is used for imaging, and light shield is arranged between optical fiber light beam light outlet and receiving screen;Method includes making filling pattern, using new spectrum point to open hole on black paper filter, obtaining spatial filter, using filling pattern to divide and fill gray scale image, and finally realizing dynamic false color display on receiving screen.With the aid of the characteristics of high resolution and fast response of spatial light modulator, digital coding sheet can be quickly switched without changing the optical path, and continuous and dynamic output of false color filling pattern is realized.

The application discloses a dual-energy XRT false color map generation method and system for photoelectric intelligent beneficiation. The method comprises the following steps: obtaining low-energy and high-energy XRT images, performing thickness reduction processing and ECC registration to obtain a ratio feature; based on the ratio feature function, fitting a material response separation curve to calculate a normalized ratio index; constructing a multi-dimensional feature vector, performing dimension reduction processing on the multi-dimensional feature vector through principal component analysis to obtain PCA principal component features; and generating a false color image by color mapping to realize dual-energy XRT false color map generation. The method combines various image preprocessing methods organically, and constructs a color mapping mechanism which has both physical interpretability and visual intuitiveness. Compared with the traditional gray-scale image method, the false color map result can more clearly distinguish ore and waste rock under complex background conditions, greatly improving the efficiency and precision of subsequent ore AI recognition algorithm training and optimization.

The application discloses a hyperspectral image efficient splicing system based on grid guidance and manual registration, relates to computer vision technology, and is proposed in view of problems such as complicated operation in the prior art. The following steps are performed during running: a hyperspectral data folder is loaded and a false color image is generated; initial image splicing is performed; a coordinate mapping file is generated and saved; a global false color image is reconstructed based on the coordinate mapping file; and a single-band gray image is generated and interactive spectrum extraction is performed. The application has the advantages that an ordered splicing mechanism of grid arrangement guidance is provided, an accurate registration strategy of manual feature point selection is combined, and independent storage and on-demand reuse of a position mapping file are deeply integrated, so that decoupling scheduling of original hyperspectral data, a false color image and a pixel mapping relationship after initial splicing is realized, splicing accuracy and operation efficiency in a complex scene are significantly improved, memory occupation of secondary reconstruction and single-band spectrum analysis is reduced, and data reuse and interactive analysis capability are improved.

A method for evaluating performance of an image signal processor according to various example embodiments includes: selecting at least two target colors from an original image generated by the image signal processor; defining a true color range in each of the target colors and identifying at least one false color pixel having a color deviating from the true color range in the original image; generating a reference image by changing a color of the false color pixel; and calculating a color error between the original image and the reference image.

The invention discloses a hyperspectral image classification method based on superpixel spectrum similarity measurement, which comprises the following steps of: extracting first three principal components to synthesize a false color image based on principal component analysis, and generating spatial coherent superpixel units by adopting an entropy rate superpixel segmentation algorithm, so that not only is the integrity of a spatial structure maintained, but also the internal spectrum information is reserved; performing edge preserving filtering processing on the original hyperspectral image by adopting domain transformation recursive filtering; spectral domain noise separation and data dimension reduction are carried out through maximum noise fraction transformation, and minimum-maximum normalization processing is carried out on all spectral features to eliminate dimension differences; the minimum spectral Euclidean distance between a test sample and each category of training samples is calculated in each superpixel unit to form a distance matrix, the most representative test pixel is screened by adopting a k-nearest neighbor average strategy to calculate a similarity score, and finally a category label with the highest similarity score is allocated to all pixels in the whole superpixel.

The application provides a kind of deep learning-based mitochondrion segmentation and classification method, solve the robustness and accuracy lower of existing technical method and model for mitochondrion morphological analysis processing, and the problem of qualitative and quantitative analysis cannot be solved, qualitative and quantitative analysis of mitochondrion super-resolution image is realized by the "segmentation-classification" idea of image segmentation and image classification combination, wherein, image segmentation part embodies quantitative, deep learning can quantify the morphological profile of mitochondrion in each image, and the fluorescence intensity of mitochondrion false color image is quantified.Image classification part embodies qualitative and quantitative, deep learning can classify mitochondrion of different morphologies to qualitatively (circular or filamentous), and the proportion of circular mitochondrion is counted to quantitatively the overall pathological degree of mitochondrion in the current cell.

The invention relates to a CT (Computed Tomography)-guided percutaneous puncture pulmonary nodule targeted tissue positioning method which comprises the following steps: acquiring perfusion images of pulmonary artery and bronchial artery, processing perfusion volume data of the perfusion images, acquiring pulmonary artery blood flow, bronchial artery blood flow and pulmonary artery perfusion indexes, and generating corresponding perfusion pseudo-color images; taking the blood supply area of the bronchial arterial blood flow as lung tumor tissues, and performing lung tumor area targeted localization according to color features in the perfusion pseudo-color graph of the bronchial arterial blood flow to obtain a target spot; the guide sheath needle based on magnetic navigation is used as a puncture positioning needle, and is positioned and punctured to lung tumor tissues according to a target spot for sampling. The accuracy and safety of pulmonary nodule pathological diagnosis are improved, complications are reduced, and a more reliable basis is provided for treatment of patients.

The application discloses a sintering machine tail section data acquisition device, method, system, equipment and medium, an infrared thermal imager is installed at the tail of the sintering machine and is used for collecting infrared images of the tail of the sintering machine; a tail multimodal sample server connected with the infrared thermal imager is used for determining a first region of interest and a second region of interest in the infrared images of the tail of the sintering machine; a first temperature value and a first image size value of the first region of interest are determined, and a second temperature value of the second region of interest is determined; a second image size value of the first region of interest in a preset sintering machine tail section false color map is acquired; if a temperature difference value between the first temperature value and the second temperature value is greater than a first preset threshold value, and a difference value between the first image size value and the second image size value is less than a second threshold value, then the acquisition time corresponding to the infrared images of the sintering machine is determined as a reference time for acquiring the sintering machine tail section data; and the sintering machine tail section data is quickly and accurately acquired based on the reference time.

The invention discloses a dual-energy XRT pseudo-color graph generation method and system for photoelectric intelligent mineral separation. The method comprises the following steps: acquiring a low-energy XRT image and a high-energy XRT image, and performing thickness cutting processing and ECC registration to obtain a ratio feature; calculating a normalized ratio index based on the ratio characteristic function and the fitted material response separation curve; constructing a multi-dimensional feature vector, and performing dimension reduction processing on the multi-dimensional feature vector through principal component analysis to obtain PCA principal component features; and generating a pseudo-color image by using color mapping, thereby realizing the generation of the dual-energy XRT pseudo-color image. According to the method disclosed by the invention, a color mapping mechanism with physical interpretability and visual intuition is constructed through organic combination of various image preprocessing methods. Compared with a traditional grayscale image method, the pseudo-color image result can distinguish the ore and the waste rock more clearly under the complex background condition, and the efficiency and precision of subsequent ore AI recognition algorithm training and optimization are greatly improved.

The invention relates to the field of endoscope imaging, in particular to a demosaicing method and system for single-channel filtering of an endoscope image. Calculating the local gradient intensity in the horizontal direction and the local gradient intensity in the vertical direction of green pixels in the color filtering array image, and performing adaptive interpolation on the green pixels according to the difference between the local gradient intensity in the horizontal direction and the local gradient intensity in the vertical direction to obtain an initial green channel; performing bilateral filtering on pixels in the initial green channel to obtain a filtered green channel; based on the filtered green channel, respectively reconstructing a red channel and a blue channel by adopting a residual interpolation method to obtain a demosaiced image; the method reduces the false color and noise amplification risk, and has the characteristics of strong edge retention capability, high overall color restoration precision, simple algorithm and easy hardware implementation.