133 results about "Infrared image processing" patented technology

The invention discloses an infrared vehicle detection adversarial patch generation method based on a particle swarm optimization algorithm, and belongs to the technical field of infrared image processing and security attack. The method comprises the following steps of: firstly, performing vehicle segmentation on an input infrared image by using an image segmentation model to obtain an accurate vehicle segmentation mask for positioning a subsequent patch; thirdly, initializing an adversarial patch according to the segmentation mask, inputting the adversarial patch into an infrared vehicle detection model (such as YOLOv5) for detection, and generating a preliminary adversarial sample; then, defining a new loss function, and optimizing the generated adversarial patch by minimizing the confidence score of a detector on the patch, so as to enhance the attack effect; and finally, iteratively optimizing the position and the shape of the patch by adopting an improved particle swarm optimization (PSO) algorithm in combination with a shape optimization module so as to maximize the attack success rate, and ensuring the robustness and the adaptability of the patch in a real environment in combination with an expectation transformation (EOT) technology. According to the method, the attack success rate of the infrared vehicle detector is remarkably improved, and an innovative and efficient solution is provided for attack resistance of the infrared vehicle detector.

The invention relates to the technical field of power failure prediction, and discloses a power equipment thermal failure monitoring and early warning system fused with infrared image processing. According to the invention, the infrared image acquisition module acquires and standardizes equipment temperature field data, and the electrical parameter synchronous acquisition module acquires load current and environment temperature in real time; the hot spot dynamic analysis module identifies and tracks an overheat area based on a dynamic threshold value, extracts spatio-temporal evolution characteristics of the overheat area, and comprehensively calculates a fault hot spot intensity factor capable of representing an early fault; the electrical fault diagnosis module starts deep diagnosis when the factor is abnormal, and accurately estimates the contact resistance value of the key part through a nonlinear recursive algorithm; and the comprehensive early warning module performs fusion evaluation on the multi-source information, and generates multi-level early warning information based on a layering strategy. According to the method, post-event alarm is converted into beforehand intervention, the sudden failure risk of the equipment is remarkably reduced, and accurate and reliable decision support is provided for predictive maintenance.

A system for monitoring a switchgear includes: an infrared camera for acquiring at least one infrared image of the switchgear; a processing unit for determining a pixel in the at least one infrared image associated with a hottest temperature, determining e pixels in the at least one infrared image associated with a temperature that is within a threshold temperature of the hottest temperature, and determining that a hot spot exists in the switchgear as a determination, the determination being a utilization of the determined pixels in the at least one infrared image; and an output unit for outputting an indication of a fault in the switchgear based on the determined hot spot.

The present application relates to the field of infrared image processing, and discloses an infrared small target detection method and device based on image information entropy and multi-scale local contrast measure, comprising: (1) constructing a local weighted information entropy operator to obtain a local weighted information entropy image of an original infrared image; (2) using the weighted information entropy image and the structure tensor theory to construct a background prior and a target prior to obtain a local saliency prior; (3) using the original thermal infrared image to design a local contrast coefficient to obtain a multi-scale local contrast image; (4) using the multi-scale local contrast image and the corresponding structure tensor to design an information filter; and (5) fusing the local saliency prior and the information filter to realize the detection of the infrared small target. The present application comprehensively utilizes the weighted image entropy operator, the multi-scale local contrast measure and the structure tensor theory, and can effectively suppress the background, enhance the target and improve the detection performance of the infrared small target.

The invention provides a gun calibration method based on automatic bullet hole detection, a medium, equipment and a product, and relates to the technical field of infrared image processing, and the method comprises the steps: capturing an infrared image of a bullet hole in real time through an infrared thermal imager, carrying out the preprocessing, and constructing a bullet hole detection data set; constructing a bullet hole detection model based on the YOLOv5 network, adding a multi-scale feature fusion module in the YOLOv5 network, and fusing the shallow-layer features, the middle-layer features and the deep-layer features extracted by the backbone network into new deep-layer features; spatial pyramid pooling is replaced by hollow spatial pyramid pooling; inputting the bullet hole detection data set into the model to obtain bullet hole position data; according to the position data of the bullet hole, converting the coordinates of the bullet hole, and calculating the offset of the bullet hole; and according to the bullet hole offset, ballistic deviation is analyzed, and ballistic deviation is corrected. The method can break through environmental limitation, realizes real-time bullet hole detection, and improves gun calibration precision.

The invention provides an unmanned aerial vehicle image ship detection method and system based on FAST features and an HSO algorithm, and relates to the technical field of infrared image processing. The problems that a traditional method cannot efficiently position the ROI and is poor in threshold segmentation effect are solved. According to the technical key points, the method comprises the following steps: carrying out FAST feature detection on a target unmanned aerial vehicle image, and determining an interest region image according to a feature detection result; determining a candidate detection threshold value set for the interest area image by adopting an HSO algorithm, performing iterative optimization processing on the candidate detection threshold value set until a termination condition is met, and outputting a detection threshold value when the termination condition is met; and carrying out image segmentation on the region of interest image by adopting the detection threshold, and carrying out ship detection on the segmented image to obtain a ship detection result in the target unmanned aerial vehicle image. According to the method, the processing range can be greatly reduced, the target detection segmentation threshold search efficiency is improved, the noise immunity is enhanced, and an image target extraction technical system adaptive to multiple complex scenes is formed.

A system for monitoring a switchgear includes an infrared camera; a processing unit; and an output unit. The camera acquires a first infrared image. The processing unit determines a pixel in the first image with a maximum temperature and uses a second number of pixels to determine a temperature interval. The processing unit sorts the pixels in descending order of temperature to determine a threshold temperature and determines that a hot spot exists in the switchgear using the temperature interval for the first infrared image. The output unit is configured to output an indication of a fault in the switchgear when the determination has been made that a hot spot exists in the switchgear.

The invention discloses a multi-scale infrared polarization image enhancement method based on MSD-PCNN, and relates to the technical field of infrared image processing, and the method comprises the steps: firstly, carrying out the multi-scale and multi-direction decomposition of an input intensity image I and a polarization feature map DoLP based on NSST, and obtaining a low-frequency sub-band and a high-frequency sub-band; secondly, according to the obtained low-frequency sub-band, designing a low-frequency sub-band fusion rule based on distance difference perception; and finally, according to the obtained high-frequency sub-band, designing a high-frequency sub-band fusion rule based on MSD-PCNN, and fully extracting texture and edge detail features. According to the MSD-PCNN-based multi-scale infrared polarization image enhancement method provided by the invention, redundant information interference is suppressed, the hierarchical definition of an image structure is improved, and meanwhile, the visual comfort and global naturalness of a fused image are also improved.

The invention relates to the technical field of infrared image processing, in particular to an autonomous inspection operation device and method for a control room, and solves the technical problem of low temperature detection precision in the prior art. The device comprises a data acquisition unit which is used for acquiring infrared image data of equipment in a control room; the infrared image data is used for representing an infrared time domain signal acquired based on infrared radiation reflection of the surface of the equipment in the control room; the signal processing unit is used for carrying out adaptive time-frequency transformation based on an infrared time-domain signal in the infrared image data to obtain a corresponding time-frequency matrix; the time-frequency matrix is used for representing energy distribution of infrared radiation on the surface of equipment in the control room at different times and different frequencies; and the inspection unit is used for separating the target heat radiation based on the time-frequency matrix and performing equipment temperature inspection according to the separated target heat radiation.

The present application relates to a kind of self-adapting wide gray scale range infrared image highlight target suppression method, belong to infrared image processing technical field.The present application utilizes the histogram information of original image, whether according to the decision function adaptive judgment enters highlight suppression algorithm processing flow according to statistical result, histogram of image is gathered and compressed in the flow, while using nonlinear mapping method, to achieve the purpose of effectively suppressing highlight target, so that highlight target and image background are effectively and reasonably displayed in display device.

The application discloses an ethylene cracking furnace tube infrared intelligent analysis system, which comprises, which are connected in sequence, an image acquisition module, a temperature monitoring module, a coking diagnosis module and a health management module; the image acquisition module acquires infrared images of cracking furnace tubes and burners by using an infrared thermal imaging online monitoring system, and establishes a cracking furnace tube infrared image database; the temperature monitoring module extracts and records the outer surface temperature of the ethylene cracking furnace tube from the processed infrared images of the ethylene cracking furnace tube by using an infrared image processing and recognition method; the coking diagnosis module combines the technologies of artificial intelligence and data mining, establishes a coking degree diagnosis and prediction model of the ethylene cracking furnace tube, and realizes early warning and targeted tracking of the coking abnormal conditions; the health management module carries out health evaluation of the cracking furnace tube according to the results obtained by the coking degree diagnosis and prediction model of the ethylene cracking furnace tube, and provides maintenance decision support. The application can realize integrated management of cracking furnace tube detection, diagnosis and maintenance.

Embodiments of the present application provide an image processing method and device for an infrared image and electronic equipment, and relate to the technical field of infrared image processing. The image processing method for the infrared image comprises obtaining a first difference image and a second difference image corresponding to a target infrared image to be processed, determining a detection result of the target infrared image on whether there is a horizontal stripe based on pixel values of each pixel point of the first difference image and the second difference image, obtaining the detection result in a time domain dimension and a space domain dimension, if the detection result in the time domain dimension and the detection result in the space domain dimension both represent that there is a horizontal stripe, performing image compensation on the target infrared image in the time domain dimension to obtain an intermediate image, performing image compensation on the intermediate image in the space domain dimension to obtain an image after removing the horizontal stripe corresponding to the target infrared image. It can be seen that the embodiments of the present application can effectively remove the horizontal stripe from the infrared image with the horizontal stripe.

The present invention belongs to the field of target recognition in infrared image processing and provides a high-time-sensitive target recognition method combined with boost segment detection to solve the problem of difficult recognition of high-time-sensitive targets. The main scheme includes: extracting the infrared radiation energy curve and space-time characteristic curve of the high-time-sensitive target, and performing outlier processing on these two types of time series signals to obtain the time series signal to be identified. Subsequently, the space-time characteristic curve is used to perform a first round of preliminary screening of non-boost segment targets. The two types of time series signals are then input into the main process of boost segment detection for a second round of screening to obtain high-time-sensitive targets containing boost segments, thereby realizing the detection of boost segments. Subsequently, key point detection is performed on the target containing boost segments to facilitate the extraction of the boost segment portion. Finally, a method of calculating waveform similarity and matching is used to classify and identify the high-time-sensitive targets containing boost segments.

The application discloses a non-uniformity correction method and system based on space domain and frequency domain features, relates to the technical field of infrared image processing, and corrects steps as follows: inputting a to-be-corrected image into a trained infrared image non-uniformity correction model to output a corrected image; model training steps are as follows: constructing a training data set based on a visible light data set; constructing an initial infrared image non-uniformity correction model, which adopts a U-Net structure to construct a non-uniformity correction backbone module, and the non-uniformity correction backbone module is provided with a space domain feature extraction module and a frequency domain feature extraction module; iteratively training the initial infrared image non-uniformity correction model by using the training data set to obtain the infrared image non-uniformity correction model. The application captures local pixel-level information through the space domain feature extraction module, combines global frequency spectrum features extracted by the frequency domain feature extraction module, realizes the synergistic effect of local noise suppression and global structure optimization, and improves the precision and robustness of non-uniformity correction.

The invention discloses an infrared video semantic segmentation method based on multi-granularity contrast learning, and belongs to the technical field of infrared image processing. The objective of the invention is to improve semantic segmentation precision without increasing reasoning time. The method comprises the following steps: carrying out time sequence consistent space enhancement transformation processing on an infrared video to obtain an annotation frame, and constructing a multi-granularity contrast learning framework by two sub-video clips with overlapped parts, namely adding three parallel multi-layer perceptron MLPs in front of a classification layer to form a context projection layer, a time sequence projection layer and a category projection layer; and on the basis of label supervision of the labeling frame, constructing a contrast learning self-supervision loss function from different granularities, wherein the contrast learning self-supervision loss function comprises a context generalization loss function, an in-video time sequence consistency loss function and a cross-video category discrimination loss function. Constructing a training set based on a difficult anchor point sampling strategy; and a trained infrared video semantic segmentation model based on multi-granularity contrast learning is obtained and is used for infrared video semantic segmentation.

The invention discloses a new energy automobile operation safety detection system and detection method, belongs to the field of new energy automobile detection, and aims to solve the problem of incomplete new energy automobile operation safety detection items in the prior art. The new energy automobile operation safety detection system comprises an OBD reading module, an electrical detection module, a direct current charging module, a charging detection module, a battery pack sealing detection module, an industrial personal computer, a host and a cloud server. According to the new energy vehicle operation safety detection system and detection method, the battery pack sealing detection module is arranged, the thermal imager and the convolutional neural network are utilized to perform sealing detection on the battery pack on the vehicle, the leakage type can be identified, and the detection accuracy is improved by utilizing an infrared image processing technology. The influence of a complex structure and a complex detection working condition is reduced, the recognition accuracy of the convolutional neural network model is improved, and then the battery pack sealing detection precision is improved.

The application discloses a kind of fan blade infrared monitoring system and method based on unmanned aerial vehicle inspection, it is related to image recognition technical field, solve the problem that the infrared image processing mode of fan blade in prior art is not enough to adapt to the temperature change of blade caused by varied environment.The infrared thermal imaging of fan blade is obtained by the infrared camera carried by unmanned aerial vehicle, the highest temperature and the lowest temperature of each infrared thermal imaging are calculated by temperature inversion, and then the temperature difference of fan blade is obtained, in addition, the infrared thermal imaging of each fan blade number in the same area is also obtained, and the highest temperature and the lowest temperature of each infrared thermal imaging are calculated, the temperature analysis set is constructed and the overall standard deviation is calculated to judge the dispersion degree of temperature in the area, and finally the health evaluation coefficient is established based on temperature difference and dispersion condition and other factors, to provide quantitative data reference for managers and decision makers, to facilitate subsequent management and decision-making work.

The application provides an infrared image processing method based on a space-time field and related equipment, and relates to the technical field of infrared image processing. The method comprises the following steps: acquiring continuous infrared images, constructing a unified background image, identifying key points of each target moving object and mapping the key points, determining an image sequence group according to a time sequence, and obtaining a complete motion trajectory curve according to the distribution of the key points. When a split-screen target infrared image is detected, the key points of each part are extracted after splitting, the two parts of the background are matched with the unified background image to determine an initial background area, the distance between the key points and the expected position predicted based on the trajectory, adjacent frames and motion rules is calculated, the current frame and the delayed frame are determined, and finally the complete sequence is obtained by processing the delayed frame. The method in the application can accurately repair the split-screen image in a dynamic scene, and ensure the time sequence coherence and integrity of the infrared image sequence.

The application relates to the technical field of edge computing, and particularly discloses an infrared image processing method for excavator operation early warning based on edge computing, which comprises the following steps: dividing an infrared image into regions according to the temperature values of pixel points in the infrared image to obtain an active region and a background region; distributing the active region to edge nodes for preprocessing according to a network condition score F of the edge nodes; after all the active regions received by an edge node j are preprocessed, the active regions are redistributed to the edge node for preprocessing; after preprocessing of all the active regions is completed, the active regions are combined with the background regions which have not been preprocessed into a target image, and the target image is used for excavator operation early warning; and the real-time degree of the excavator operation early warning is improved.

The present invention provides a dual-camera face recognition method without a white light fill light, comprising the following steps: acquiring a real color image and an infrared image; processing an experimental color image; inputting the experimental color image and the infrared image for training, and outputting a fused image; calculating a loss function and adjusting model parameters; starting a detection program, waiting for the light illuminance value to stabilize, acquiring the current light illuminance value and white balance value, and switching between a bright field mode and a dark field mode when the light illuminance value and the white balance value meet a switching condition; if the bright field mode is currently in the bright field mode, performing face recognition based on the color camera and the infrared camera; if the dark field mode is currently in the dark field mode, collecting a target color image and a target infrared image in the current environment, inputting them into an image fusion model, obtaining a target fused image, and performing face recognition based on the target fused image and the infrared camera. The present invention also provides an electronic device and a medium for solving the problem of people not being dazzled by white light fill lights at night.

This invention relates to the field of infrared image processing technology, and more particularly to an infrared image processing method and system based on artificial intelligence. The system includes an infrared image processing center, a scene acquisition module, an image classification module, an image enhancement module, a processing fusion module, a task selection module, and a visual output module. It classifies the original infrared image using a task recognition sub-model, accurately determining three types: conventional processing, mixed defects, and high-probability defects. Mixed defect images are enhanced using a multi-attribute weighted fusion method, while high-probability defect images undergo specific processing according to scene priority, ensuring that core defects are addressed effectively and avoiding a one-size-fits-all approach, significantly improving the targeting of image processing. Simultaneously, by dividing the target area and background area, it objectively assesses the quality from three dimensions: information retention, detail clarity, and noise variance, providing intuitive feedback on whether the processing results meet standards, facilitating timely adjustments to processing strategies by management personnel.

The invention relates to the technical field of infrared image processing, and discloses an infrared image enhancement method based on radiation priori and a generative adversarial network, which comprises the following steps of: constructing a training data set of non-paired low-quality infrared images and high-quality infrared images, constructing a bidirectional cyclic generative adversarial network, pre-establishing infrared radiation quantity databases of different materials, and constructing a training data set of the non-paired low-quality infrared images and high-quality infrared images; the method comprises the following steps: adding a material coding branch in a generative network, inputting the material coding branch in a mapping database, obtaining radiation coding characteristics, constructing a composite loss function, and carrying out feedback optimization on a discriminant network and the generative network by using a training data set through the composite loss function to obtain a trained overall enhanced network. The infrared image to be enhanced is input into the enhancement network, the enhanced infrared image is output, and the defects that an existing infrared image enhancement method depends on paired data, material radiation characteristics are ignored, enhanced image semantics are inconsistent and the like are overcome.

This invention relates to a method and system for super-resolution reconstruction of infrared images based on a micro-scanning device, belonging to the field of infrared image processing technology. The method includes: controlling a micro-scanning device to acquire multiple frames of low-resolution infrared images with sub-pixel displacement; constructing a high-resolution grid, projecting pixel values ​​onto the high-resolution grid through sub-pixel fusion, and accumulating weights to generate a reconstructed image and a weight matrix; generating a confidence map based on the weight matrix, calculating an adaptive sharpening coefficient based on the confidence map, and injecting high-frequency details into the reconstructed image to obtain a 2x super-resolution reconstructed image. Further, a dual-cascade fusion architecture is adopted, transforming the original multi-frame images into four sets of images, each reconstructed at 2x, and then performing a secondary fusion on the resulting intermediate images to achieve 4x super-resolution reconstruction. This invention can effectively suppress artifacts in non-ideal sampling areas, overcome the magnification limitations of fixed-magnification devices, and is independent of training data, exhibiting strong generalization ability.

The application discloses a kind of pre-processing method and system for rock fissure infrared image analysis, it is related to infrared image processing technical field, the method includes: the noise level estimation is carried out to the original infrared image of rock fissure obtained, generates noise feature description information;Based on noise feature description information, the original infrared image is carried out adaptive denoising processing, generates denoising image;Based on denoising image, background separation and high-frequency component extraction are carried out, and generate high-frequency feature image;Based on high-frequency feature image, carry out detail enhancement processing, and generate detail enhancement image;Based on detail enhancement image, carry out adaptive brightness correction processing, and generate pre-processing output image.Adaptive noise suppression and edge preservation to rock fissure infrared image are realized, fissure high-frequency detail feature and trend continuity are strengthened, and the contrast of low brightness area is improved, and high-quality data basis is established for downstream detection and segmentation task.

The application relates to a kind of double-band based infrared image blind cell compensation methods, belong to double-band infrared image processing technical field.The method includes: the double-band infrared image of input X i ( i , j )、 Y i ( i , j ) respectively carry out blind cell detection;Edge extraction is carried out to the double-band infrared image, and edge image Ex i , j 、 Ey i , j ;According to the detection condition and extraction condition, the pixel condition of the double-band infrared image is classified, and the image is output after the pixel is compensated according to the classification condition.The application method can improve the accuracy of blind cell detection, improve the accuracy of blind cell compensation, and effectively improve the compensation effect of large blind cell area in the image by using the difference of double-band background radiation information.​​

The present disclosure provides an image processing device, an imaging apparatus, an image processing method, and an image processing program capable of accurately extracting an abnormal region of a subject from an image. In the image processing device according to one aspect of the present invention, a processor generates a first mask image showing a first region where an abnormality exists on a surface of a subject and a second region including a region where an abnormality exists inside the subject, based on a visible image and a first near-infrared image, acquires a second near-infrared image of the subject captured with light in a second wavelength range in a state where the subject is held in focus with the light in the second wavelength range, and extracts an abnormal region in the first region and the second region based on the first mask image and the second near-infrared image.

The invention relates to the field of infrared image processing, and discloses an AI automatic distance measurement method based on an infrared image model, which comprises the following steps: deploying a high-precision thermal infrared imager, and collecting multi-scene and multi-distance infrared images; preprocessing the original infrared image; establishing a data set containing a plurality of typical scenes according to the preprocessed image, and marking the image; an improved yolov5 model is adopted to train the data set, and accurate target detection frame information is obtained; sampling different environment temperature image samples and corresponding actual physical distance data to form a ranging training sample; inputting the accurate target detection frame information and the distance measurement training sample into a designed Gaussian radial basis function neural network model for training to obtain a physical distance measurement model, and compensating the error of the system by using the physical distance measurement model to obtain the distance between the target and the thermal infrared imager; the method has the beneficial effects that the ranging precision is improved, the environmental adaptability is enhanced, the real-time performance is excellent, and the deployment is simple and convenient.

This invention discloses an infrared anomaly region detection and segmentation method based on a cascaded architecture, belonging to the fields of infrared image processing and computer vision technology. Addressing the problems of high false detection rate and low segmentation accuracy in existing infrared anomaly region detection methods, this method employs a three-stage cascaded detection-filtering-segmentation architecture: After preprocessing the infrared image, it is input into a target detection model to generate candidate bounding boxes; multi-bounding box fusion technology is used to calculate spatial cue confidence; a secondary confidence assessment and false positive filtering are performed based on the channel attention mechanism and inverse residual structure of a lightweight feature extraction network, generating a filtering score; high-confidence detection boxes are converted into effective spatial codes and used as spatial cue words input into the segmentation model to complete pixel-level fine segmentation. This invention improves positioning accuracy through a multi-bounding box fusion confidence mechanism, reduces the false detection rate through secondary filtering, and achieves high-precision pixel-level segmentation without manual annotation through effective spatial code conversion.

A medium voltage switchgear or controlgear monitoring system includes: an infrared camera; and a processing unit. The infrared camera is mounted within a medium voltage switchgear or controlgear. The infrared camera is configured to acquires an infrared image including image data of two or three current carrying parts of the switchgear or control gear. The two or three current carrying parts are the same current carry part of two or three equivalent systems within the switchgear or controlgear. The infrared camera provides the infrared image to the processing unit. The processing unit determines that the two or three current carrying parts are operating correctly or that one of the two or three current carrying parts has a fault. The determination includes analysis of the infrared image by an autoencoder implemented by the processing unit.

The application discloses a pedestrian infrared image segmentation method and system under complex conditions, applied to the technical field of infrared image processing, comprising: using the detail information in the visible light image, and adopting a guided image filtering method to carry out denoising processing on the collected pedestrian infrared image, to obtain a processed image; an improved pulse coupled neural network model is constructed, comprising: taking the background score of each pixel in the image as the connection strength, and taking the clustering center of the background area of the image as a dynamic threshold; the improved pulse coupled neural network model is used for image segmentation on the processed image, to obtain a segmented image. In this way, the improved model obtained by the application can automatically stop according to the segmentation result, and over-segmentation phenomenon is avoided, so that the infrared pedestrian images in various complex environments can be effectively segmented, and the infrared images with more heat source interference in the background have a good segmentation effect.