153results about How to "Precision Fusion" patented technology

The invention discloses a dual-CCD (Charge Coupled Device) scanning imaging detection method for continuous casting slab surface defects. The dual-CCD scanning imaging detection method comprises the following steps: (1) mounting a linear array CCD camera, an area array CCD camera and a linear laser transmitter on a continuous casting slab; (2) scanning by the linear array CCD camera and the area array CCD camera according to a synchronous frame frequency and line frequency manner to obtain a two-dimensional image and a three-dimensional image of the surface of the continuous casting slab; (3) registering the acquired two-dimensional image and three-dimensional image; and (4) carrying out information fusion processing on the registered two-dimensional image and three-dimensional image in one area on the surface of the continuous casting slab, analyzing area-of-interest features of the two images, and extracting a final area-of-interest to be identified on the surface of the continuous casting slab by combining the feature fusion processing result of the two images, thus reserving reliable defect information of the surface of the casting slab and detecting the surface defects of the continuous casting slab. According to the dual-CCD scanning imaging detection method provided by the invention, the structure is simple and convenient; the method is high in detection precision, great in real-time property and capable of automatically eliminating interference information.

The invention discloses a bus road condition processing system and method based on traffic multi-source data fusion. The bus road condition processing system comprises a preprocessing module, a source data processing module, a data source evaluation module and a data fusion module. The preprocessing module is used for processing traffic multi-source data; the source data processing module is used for processing data of a non-bus data source road condition calculation result; the data source evaluation module comprises an evaluation framework used for evaluating the quality of data sources, and the credibility of multi-source data to the bus road condition calculation result can be obtained through processing of the evaluation framework; the data fusion module is used for performing weighted calculation on the multi-source data based on the data source evaluation result, and finally the required bus road condition data are obtained; a display module is used for superposing a bus road condition map and a map file obtained through processing and displaying road condition information of the whole bus road network. By synthesizing the multiple traffic data source road condition calculation results, the advantages of all the sides can be made the best of, the fusion result information used for describing the bus road conditions can be obtained, and data support is provided for bus service.

A method and apparatus for reduction of ghost images in stereoscopic images. This disclosure provides a ghost compensation apparatus and methods that detect affected regions where ghosting may occur in a stereoscopic image, yet where conventional ghost compensation techniques are ineffective because there is insufficient luminance overhead to conduct a conventional ghost compensation process. Luminance values are modified in such regions prior to applying a ghost compensation process.

The invention relates to a method and a device for true color 3D object holographic display. Colorful image information of a true color 3D object is decomposed into RGB coloring information which comprises component amplitude information and depth information of each color, and a multi-component-level phase type kinoform of each color component of the 3D object is calculated according to the depth information and amplitude information of each color component; image series formed by each color component kinoform are output respectively to three spatial light modulators with phase modulation capacity under the control of an isochronous controller, simultaneously relevant light of relevant color is adopted to illumine the kinoform carried by three spatial light modulator, beams modulated by the kinoform show the component information of RGB of an original true color 3D object, and holographic true color reappearance effect is obtained after integration of the component information. The invention designs a plurality of devices and provides a position relation setting method of key devices in an opto-electrical reappearance device. The invention can be used for holographic display of a static or dynamic true color 3D object or scene.

The invention discloses a data processing and fusion method for large-scale traffic information and belongs to real-time traffic information processing technologies. The method includes: obtaining normal data of a truth value according to multisource traffic data acquired by a test vehicle and each sensor, and determining a dynamic assignment method for parameters; rejecting abnormal data in data sets acquired by the sensors, and performing historical data compensation; and fusing real-time grading information of the multisource traffic data with data compensation completed. The truth value system is set up for obtaining correct initial assignment parameters of various acquisition modes and rejecting acquired abnormal data, missing data are reasonably filled according to the historical data to guarantee accuracy and integrity of the data, and the data acquired by different acquisition modes are graded and fused gradually to guarantee reliability of the data. Moreover, quickness in the fusion process is achieved, and influences of traffic incidents, traffic control, land occupation construction and traffic accidents on data in the fusion process are considered.

The invention discloses a system for integrated management of production and operation of a city pipe network. The system for integrated management of production and operation of the city pipe network comprises a GIS, a dispatching center SCADA system and a full pipeline monitoring and detecting system, wherein monitoring data of all stations are sent to the GIS through the full pipeline monitoring and detecting system, the GIS sends the data to the dispatching center SCADA system, and then real-time monitoring and dispatching management of the pipe network are achieved. By the adoption of the system for integrated management of production and operation of the city pipe network, integrated management of the spatial data, the attribute data and the topological relation of pipelines is achieved, all-weather operation, real-time monitoring and real-time alarming are achieved, the multimedia information integration function is achieved, high integration and efficient infusion of production dispatching and all systems are achieved, and all kinds of information can be displayed accurately and shared highly.

The invention discloses an infrared and visible light image real-time fusion system based on a heterogeneous multi-core architecture. The system uses a heterogeneous multi-core processor, with an ARM core and a DSP core as cores, to take charge of the control flow and the data sending and receiving and processing logic of the whole system, complete video data collection and real-time enhancing and fusion of infrared and visible light images, and display and output the video data after fusion. The ARM core is responsible for real-time data collection of both infrared light and visible light video data and real-time transmission of compressed data after image fusion. The DSP core is responsible for registration preprocessing and image enhancement preprocessing of the infrared light input image, real-time fusion of the infrared light and visible light video data, and real-time encoding compression of the video data after fusion. A three-layer Laplacian pyramid structure is used for the real-time fusion. The system is high in computation efficiency, good in compatibility, high in extensibility and adaptability, and convenient and highly efficient in software developing.

The invention relates to an ROS-based laser radar and camera fusion calibration system and calibration method, and the system comprises: a radar calibration module which is used for obtaining the external parameters of a laser radar, i.e., the conversion relation from a radar coordinate system of the laser radar to a world coordinate system; a camera calibration module which is used for acquiringinternal parameters of the camera; a fusion calibration module which is used for receiving the external parameters of the laser radar and the internal parameters of the camera and solving the externalparameters of the camera, namely solving a transformation relation from a world coordinate system to a camera coordinate system. Data transmission among the radar calibration module, the camera calibration module and the fusion calibration module is realized based on an ROS platform. the radar calibration module comprises a laser radar, a pile barrel and two calibration plates, and the camera calibration module comprises a camera and a calibration plate. According to the space synchronization method for realizing data fusion of the laser radar and the camera, the defects of insufficient sensing dimension, lack of information and the like existing in the conventional single sensor sensing are overcome.

The invention, which belongs to the field of digital image processing, particularly relates to a face editing and complementing method and device based on geometric face information so that a technical problem of realizing precise complementing and attribute editing of a missing face picture can be solved. The face editing and complementing method comprises: on the basis of a pre-built face geometric estimation model, geometric face information is obtained by using a target face image including a shielded region; according to a pre-constructed face completion network model, a first complete face image is generated by using the target face image and geometric face information; and a specific area in the first complete face image is replaced with a non-shielded region in the target face image and image fusion is carried out on the first complete face image and the non-shielded region to obtain a second complete face image. With the method provided by the invention, a missing face picturecan be complemented and edited precisely.

The invention discloses a visual speech multi-mode collaborative analysis method based on an emotion context and a system. The method includes (S1) dynamically extracting and analyzing emotion context information based on a situation and an analysis object in a visual speech situation; (S2) extracting visual emotion characteristics of the analysis object in the visual situation and speech emotion characteristics of the analysis object in a speech situation in real time; (S3) carrying out structural sparse representation for the emotion context information, posture characteristics, expression characteristics and the speech emotion characteristics; and (S4) carrying out collaborative analysis and identification for the multi-mode emotion information by means of sentiment classification agents. Due to the facts that the emotion context information, the posture characteristics, the speech emotion characteristics and the expression characteristics include large amounts of emotion information, the emotion information is complementary, and the structural sparse representation and collaborative analysis through multiple sentiment agents are combined, emotion of a to-be-analyzed person is precisely analyzed on the condition that partial channel information is lost, and precision and robustness of emotion analysis in the natural interactive environment can be improved.

The embodiment of the invention provides a three-dimensional scene fusion method and device based on monocular estimation. The method comprises the steps: inputting an obtained first image into a target monocular depth estimation network, and obtaining a target depth map, the target monocular depth estimation network being obtained through the training of an initial monocular depth estimation network; obtaining depth information of a target object in the target depth map according to the target depth map and the target semantic segmentation map; according to the depth information of the targetobject and the parameter information of the equipment for collecting the first image, the position information of the target object in a preset static three-dimensional scene is obtained, and a mapping relation exists between a coordinate system of the static three-dimensional scene and a world coordinate system where the target object is located. The problem that in the prior art, due to the fact that a target depth estimation method is not ideal in implementation, the fusion mode of a monitoring object and a static three-dimensional scene model is not ideal is solved.

The invention discloses a remote sensing image fusion method based on a dual-channel neural network, and the method comprises the steps of firstly, extracting the spatial detail information in a high-frequency component after the high-pass filtering of a panchromatic image through the joint learning of a dual-channel network by utilizing the ARSIS thought; then injecting spatial detail informationinto each waveband image of the multispectral image by using a detail injection model to obtain a required high-resolution multispectral image. According to the method, the advantages of deep learning are effectively utilized, the spatial detail information, different from multi-spectral image waveband images, of a panchromatic image is obtained through double-channel network joint training, andthe correlation between the interior of the image and the images is fully utilized, so that the detail reconstruction is more accurate. Meanwhile, the number of injected details is effectively controlled by using the detail injection model, so that the spatial information storage performance of the fusion algorithm is remarkably improved, and the spectral characteristics of the original multispectral image are well kept.

A fuser for fusing an image to print media in a color electrophotographic printer. The fuser includes an endless idling belt defining an inner loop, and a ceramic heater positioned in contact with the belt, within the inner loop. A pressure roller defines a nip with the belt. The belt includes a compliant layer for conforming to variations in toner pile height. The heater is configured to provide a cooler nip exit and a hotter nip entrance.

The invention relates to an nvoice seal removing method, device and computer storage medium. The method comprises the following steps: collecting invoice image samples containing seal patterns, forming a data set, training the data set to obtain a trained detection model; detecting an invoice image to be sealed by using the trained detection model, extracting a seal region image in the invoice image, and determining a color of the seal pattern; Removing the seal in the seal area image according to the color of the seal pattern to obtain a seal area background image; the background image of theseal area is fused with the invoice image after the seal area image is extracted to obtain a target invoice image with the seal eliminated. The method for eliminating the invoice seal of the invention can better eliminate the seal, avoid influencing the identification of other text information in the invoice, obviously improve the accuracy of the invoice identification and improve the working efficiency of the invoice identification.

A combined navigation method based on an INS (inertial navigation system) / GPS (global position system) / SAR (synthetic aperture radar) comprises steps as follows: an INS receiver, a GPS receiver and an SAR sensor are mounted in appropriate positions of an unmanned aerial vehicle in parallel; image matching is performed, and position navigation deviation of the INS is obtained and taken as an observed quantity to be input into a filter, and the observed quantity and other observed quantities are jointly filtered; GPS measuring deviation is obtained and taken as an observed quantity to be input into the filter, and the observed quantity and other observed quantities are jointly filtered; the filter integrates the measuring deviation from the GPS and the position navigation deviation, obtained through image matching, from the INS, and a navigation error estimated value is calculated. Topographic features are automatically and reliably extracted from an SAR image, a map image is formed in a digital map, and the SAR topographic features are automatically predicted; topographic deviation is accurately estimated and integrated; capture and matching are automatically initialized; the fault tolerance capability, the autonomy and the reliability of navigation are high.

The invention provides a seismic property combination method based on a Dempster / Shafe (D-S) evidence theory. According to the seismic property combination method based on the D-S evidence theory, source property data undergoes feature extraction and evidences are formed to obtain basic probability values. Under instruction of the D-S evidence theory, the basic probability values are combined to obtain a combined probability value. The combined probability value is used as a weighting coefficient and furthermore, all partial small blocks of attribute data are combined with a weight number to achieve combination of multiple seismic properties. Therefore, when single seismic property information is used to explain a complicated geology problem, a problem that multiple solutions are produced by the single seismic property information is solved.

The invention relates to the digital image processing field and the image fusion algorithm field, for greatly improving the running speed of the whole fusion system on the premise that the quality and detail of the image are not impacted, the data volume of the image is greatly reduced and the amount of the data to be processed for fusion process is promoted. Therefore, according to the technical scheme, the image fusion method of compressed sensing framework comprises following steps: step 1, executing the single layer wavelet decomposition on the image to be fused; step 2, selecting the Gaussian along matrix as the measure matrix, measuring and obtaining the value; step 3, selecting the numerical value corresponding to each point for forming the new matrix; step 4, adopting the principal component analysis method for fusing the high-frequency components of the source images after measured; step 4, obtaining the high-frequency component, which is suitable for the wavelet inverse transform; and step 5, obtaining the fused image. The image fusion method of compressed sensing framework is mainly used for the digital image processing field.

The invention discloses a DRNN neural network-based eggplant greenhouse temperature intelligent detection device. The detection device is characterized in that the detection device is composed of a CAN bus-based eggplant greenhouse environment parameter acquisition platform and an eggplant greenhouse temperature intelligent detection system. An existing eggplant greenhouse monitoring system failsto detect the temperature of the greenhouse environment of eggplants according to characteristics such as the non-linearity and large lag of the temperature change of the environment of an eggplant greenhouse and complex temperature change due to the large area of the eggplant greenhouse, and as a result, the adjustment and control of the temperature of the environment of the eggplant greenhouse is greatly affected, while, with the DRNN neural network-based eggplant greenhouse temperature intelligent detection device adopted, the above problem can be solved.

The invention discloses a medical image fusion method based on an improved pulse coupling neural network. The medical image fusion method comprises the following steps of 1, performing Gamma correction on a multi-mode medical image to enhance the contrast ratio of the medical image; 2, performing multi-scale decomposition on the corrected image to be fused by adopting non-subsampled shear waves toobtain a low-frequency subgraph and a high-frequency subgraph; 3, fusing the low-frequency sub-graphs by adopting an improved regional energy algorithm; 4, fusing the high-frequency sub-graphs by adopting an improved pulse coupling neural network algorithm; and 5, reconstructing the fused high-frequency subgraph and the fused low-frequency subgraph by adopting non-subsampled shear wave inverse transformation to obtain a final fused image. The multi-modal medical image fusion method can effectively fuse multi-modal medical images, and improves the accuracy of doctors in diagnosing the illnessstate of a patient.

Disclosed is a data processing system and method for the fusion of multiple heterogeneous sensors. The data processing system and method may synchronize the first merged sensing signal with the second sensing signal by performing origin time shifting (OTS) and tracklet movement compensation (TMC). Namely, it is possible to perform time synchronization to the heterogeneous sensors by post-processing data without synchronizing the data acquisition time points through a trigger. Also, the data processing system removes the noise points caused by direct and indirect interference included in the first sensing signal, thereby providing more accurate sensing signal data.

The present invention provides an adaptive scale image sequence target tracking method based on feature filtering and fast motion detection template prediction. The method comprises the following steps of: 1, specifying a to-be-tracked object in an initial frame, transforming the object into a frequency domain by fast Fourier transformation based on a Gaussian kernel, and constructing a feature filter; 2, constructing a fast motion prediction template; 3, constructing an object imaging scale change model; 4, in a second frame, acquiring a predicted position of the object by the fast motion prediction template, acquiring a position of a corresponding matched search box according to the object imaging scale change model by using the predicted position as the center, performing filteration by the feature filter, regarding a position of a maximum response value as an object location, and simultaneously, generating a new feature filter and using the new feature filter as a to-be-used feature filter in a next frame; 5, sequentially repeating the step 4 in subsequent frames until a video or image sequence is finished. The method provided by the present invention is effectively applicable to tracking of a distant and small object in a complicated background and has good tracking performance.

The present disclosure discloses an image processing method, including: recognizing basic feature points of a source object in a source image and basic feature points of a base object in a base image; determining, according to distribution of the basic feature points in each of the images, auxiliary feature points that meet a filling condition according to the distribution in a corresponding image; determining feature points of a same type in the source object and the base object, the feature points of the same type including basic feature points of a same type and auxiliary feature points of a same type in the source object and the base object; determining average feature points according to positions of the feature points of the same type; deforming the source object and the base object according to the average feature points; and fusing the deformed source object with the deformed base object. The present disclosure further discloses an image processing apparatus, a terminal, and a storage medium.

The invention discloses a camera calibration method of a three-dimensional laser imaging system. The method utilizes characteristics points to establish a camera coordinate system, then obtains six deviation parameters of the camera coordinate system and a carrier coordinate system through differential computing, completes the calibration of the camera coordinate system, and finally obtains the precise conversion matrix from the carrier coordinate system to the camera coordinate system. At present no corresponding method can process the camera calibration method of a three-dimensional laser imaging system. The method provided by the invention is simple and clear, the programming is easy to realize, and the precision of integration of point cloud data and pictures in a three-dimensional laser imaging system is improved.

The invention discloses a surgical microimaging system and method based on optical coherence tomography augmented reality. The system comprises a surgical microscopic unit which is used for collectingtwo-dimensional microscopic images of a surgical region, an optical coherence tomography unit which is used for collecting OCT three-dimensional images of the surgical region, a guiding light sourcewhich can be captured by a camera of the surgical microscopic unit and is used for projecting a guiding light spot synchronized with an OCT scanning light source of the optical coherence tomography unit, a processing control unit which is used for acquiring the two-dimensional microscopic images and OCT three-dimensional images of the surgical region, as well as images after the two-dimensional microscopic images and the OCT three-dimensional images of the surgical region are fused, and a display unit which is used for outputting and displaying results of the control unit. The surgical microimaging system can be used for accurately registering and fusing two-dimensional microscopic images and OCT three-dimensional images, realizes microscopic image real-time augmentation of microscopic images of the surgical region, provides intuitive navigation information to surgery, and realizes intuitive surgical guidance.

The invention discloses a mobile robot map construction method based on visual inertial navigation fusion and related equipment, and the method comprises the steps: S1, determining a fusion achievement degree according to a visual attitude and inertial attitude fusion state, and completing visual initialization, visual inertial calibration alignment and inertial initialization; s2, constructing a situation evaluation function to analyze a posture trend in real time; s3, determining a fusion state local expected value and outputting the state; s4, dividing the key frame into a strong common-view key frame and a weak common-view key frame according to the point sparse degree of the public map; and S5, sending the strong common-view key frame into a sliding window according to a grading result, optimizing the strong common-view key frame by utilizing inter-frame pose dense continuity, and drawing and correcting the self-sensing map by virtue of a relation of observation map points. Accurate fusion between the camera data and the IMU data is guaranteed, and the system is high in efficiency of reading the current map point pose information and small in error.

The invention relates to a high-voltage insulation fault diagnosis method based on heterogenous image temperature rise and partial discharge characteristics. The high-voltage insulation fault diagnosis method based on heterogenous image temperature rise and partial discharge characteristics includes the steps: extracting the contour of an infrared image and the contour of an ultraviolet image; performing image registering based on the contour information to obtain the optimal affine transformation parameter; by means of the optimal affine transformation parameter, performing ultraviolet imagetransformation, extracting the ultraviolet spot contour diagram and the ultraviolet gray level image of the ultraviolet image after transformation, and performing fusion of the infrared image and theultraviolet image; and extracting the greatest temperature rise and the partial discharge characteristics from the fused image, and at the same time constructing a fuzzy logical reasoning system to perform external insulation fault diagnosis. Compared with the prior art, the high-voltage insulation fault diagnosis method based on heterogenous image temperature rise and partial discharge characteristics has the advantages of reducing the information redundancy, being more visual in the images, improving the detection accuracy, being stable in the algorithm, being high in the information reservation degree, being high in the applicability and the like.

The invention relates to the field of digital image processing, and provides a method capable of effective showing and processing images and other higher spatial data based on compressed sensing and WBCT. The method comprises the following steps that firstly, small-wave based Contourlet transformation is carried out on a to-be-fused image; secondly, a measurement matrix is formed, and measuring is carried out on LH, HL and HH high frequency sub-bands; thirdly, pixel point leveled evaluating is carried out on low frequency components of two source images by the rule adopting the Weighted Average (WA); fourthly, fusion is carried out on measured high frequency components of the two source images by using the principal component analysis; fifthly, reconstitution calculation is carried out on the fused high frequency components by using StOMP; sixthly, inverse transformation is carried out by using WBCT. The method is mainly applied to digital image processing.

The invention discloses a multi-functional headset intelligent fire fighting system and a control method thereof. According to the multi-functional headset intelligent fire fighting system, a vibration-resistant platform is adopted to eliminate high-frequency vibration in the movement process of a fireman, an ultrasonic sensor makes up for the deficiencies of an infrared camera on deep information perception, and three-dimensional physical environment and temperature field distribution in a fire scene environment are calculated through a fusion algorithm; meanwhile, a combined sensor of GPS and inertial measurement and a magnetic field sensor are adopted, inertial movement information, magnetic field information and position speed information are fused, and information, such as a position, a speed, a posture and the like, relative to a node can be accurately acquired, so that a position and a direction of the node in a whole fire scene is assisted to be accurately established, and the three-dimensional physical environment and the temperature field distribution in the whole fire scene environment is established accurately by integrating various nodes; and in addition, by combining with historical data, optimal action strategy is obtained. The multi-functional headset intelligent fire fighting system accurately perceives the fire scene environment, rapidly analyzes a form of a fire and can perform intelligent decision, and thus own protection and efficient rescue and fire-fighting of the fireman is realized.