17 results about "Tomographic image reconstruction" patented technology

The invention discloses a multi-frame photoacoustic image reconstruction method based on optical flow alignment and depth feature fusion. The method comprises the following steps: S1, obtaining photoacoustic signal data; s2, reconstructing a photoacoustic cross-sectional image; s3, performing optical flow calculation and image alignment; and S4, training the deep feature fusion network. According to the method, through optical flow motion correction and a potential space learning mechanism, space-time information and complementary features in the aligned multiple frames of images are dynamically integrated, complementary information in the multiple frames of images is adaptively fused, noise is suppressed, and finally reconstruction of high signal-to-noise ratio and high spatial resolution images of biological tissues is achieved. Experimental results show that the method can significantly improve image quality, recover image distortion and detail loss caused by motion and noise, and provide a new effective scheme for promoting robust clinical application of a photoacoustic imaging technology.

The invention discloses a complex metal product internal defect analysis method and system based on radiographic inspection, a spiral track scanning path is generated through a workpiece three-dimensional model, a six-axis cooperative mechanical arm is utilized to drive a miniature X-ray source to move along the path, and + / -5 [mu] m positioning compensation is realized in combination with a piezoelectric ceramic driving mechanism. And the detector array module is matched to carry out multi-mode ray scanning and real-time data acquisition. Collected data are transmitted to a real-time data processing module through an LVDS differential interface, an FDK algorithm is executed by a JetsonTX2 embedded GPU to complete tomographic image reconstruction, a 3D-ResNet neural network model is loaded through a TensorRT inference engine to achieve intelligent defect recognition, and finally a detection report is generated and uploaded to an MES system. The system comprises a ray generation module, a multi-degree-of-freedom mechanical arm bearing module and the like, a high-voltage generator circuit comprises a ZVS resonance circuit, and a motion control module adopts an FPGA + ARM dual-core architecture.

The application relates to the technical field of medical image reconstruction, and discloses a three-dimensional ultrasonic tomographic image reconstruction method based on time-domain full waveform inversion, which comprises the following steps: establishing a time-domain wave equation of ultrasonic wave propagation in biological tissues, generating a simulated sound pressure distribution according to the time-domain wave equation; using a mean square error to measure the error between a real sound pressure distribution and the simulated sound pressure distribution, using an adjoint field method to calculate the functional gradient of the mean square error, and using a momentum gradient descent method to minimize the error; in the process of minimizing the error, reconstructing a sound velocity distribution according to the functional gradient of the error with respect to the sound velocity, then using the reconstructed sound velocity distribution to reconstruct a sound absorption coefficient distribution, and repeatedly and alternately reconstructing until a set condition is reached, so that the reconstruction of the sound velocity distribution and the sound absorption coefficient distribution is completed. The application simultaneously reconstructs the sound velocity and the sound absorption coefficient in the time domain, avoids the problem that a traditional full waveform inversion algorithm needs to reconstruct the sound velocity in the time domain and reconstruct the sound absorption coefficient in the frequency domain, and reduces the algorithm complexity.

Example methods and systems for tomographic image reconstruction are provided. One example method can include obtaining two-dimensional (2D) projection data (310) and processing the 2D projection data using an AI engine (301) that includes a plurality of first processing layers (311), an interposed back-projection module (312), and a plurality of second processing layers (313). Example processing using the AI engine can include generating 2D feature data (320) by processing the 2D projection data using the plurality of first processing layers, reconstructing first three-dimensional (3D) feature volume data (330) from the 2D feature data using the back-projection module, and generating second 3D feature volume data (340) by processing the first 3D feature volume data using the plurality of second processing layers. Methods and systems for tomographic data analysis are also provided.

The invention relates to a computed tomography image reconstruction method and system based on extremely sparse scanning. The method comprises the following steps: acquiring original scanning data of a CT machine and a reconstruction image of a standard algorithm; building a generative adversarial network; designing a loss function, and training the generative adversarial network; a VdCS algorithm and a CT-RDNet network are combined through a feedback path, and an extremely sparse scanning CT image reconstruction framework, namely a CSUF image reconstruction framework, is constructed; acquiring scanning data, namely extremely sparse scanning data, of an extremely sparse scanning mode of the CT machine; and taking the extremely sparse scanning data as input, executing extremely sparse scanning CT image reconstruction by the CSUF image reconstruction framework, and outputting a reconstructed image meeting imaging clinical diagnosis requirements. According to the method, the CT image meeting the imaging clinical diagnosis requirement can be obtained under the extremely sparse scanning condition, and good robustness and engineering practicability are achieved.

The invention provides a metallurgical mineral conveying process component real-time monitoring method and a metallurgical mineral conveying process component real-time monitoring system based on electromagnetic tomography. The method comprises the following steps: carrying out non-contact scanning on a conveyor belt by adopting a multi-frequency excited and eccentrically arranged EMT sensor array, and constructing a mineral-electromagnetic characteristic mapping database in combination with an electromagnetic tomography image reconstruction algorithm and a deep learning method; real-time mineral pixel-level identification is carried out through the trained machine learning model; the recognition result is divided according to grid areas, and the real-time mass ratio data of all minerals on the material section are calculated in combination with the pre-calibrated mineral density and the area weight factor corrected by the belt curvature; and signal drift compensation is carried out by fusing high-precision temperature sensor data. And finally, real-time mass ratio data are converted into control signals, downstream equipment such as a sorting mechanical arm and a batching valve is driven, online recognition, precise quantification and intelligent regulation and control of mineral components are achieved, and the resource utilization efficiency and the automation level in the metallurgical process are remarkably improved.

The application belongs to the technical field of photoacoustic imaging, and discloses a unipolar adaptive photoacoustic tomographic image reconstruction method, which comprises the following steps: firstly, inserting a virtual array element between adjacent array elements of an ultrasonic array, and improving the spatial sampling rate through interpolation; secondly, reconstructing an original photoacoustic image based on a delay-and-sum algorithm; thirdly, performing time difference processing on the interpolated channel signals to obtain first-order time gradient data, and introducing a linear compensation factor to correct the negative gradient signals; fourthly, reconstructing a gradient photoacoustic image by using the compensated gradient signals; and finally, normalizing and then weighting and fusing the original image and the gradient image to generate a high-contrast unipolar photoacoustic image. The application compensates for negative signals through mathematical operation and fuses images, effectively solves the problems of low contrast and fuzzy edges of a traditional bipolar image, and avoids the loss of details caused by Hilbert transform. On the basis of retaining the characteristics of the sound pressure signal, the method significantly improves the image signal-to-noise ratio and edge sharpness.

The invention relates to an emission tomography image reconstruction method. The method comprises the steps that an emission tomography data set is acquired, iterative reconstruction is performed on the emission tomography data set to obtain a target reconstruction image, and in the iterative reconstruction process, the following steps are performed in at least one iteration: an intermediate iteration image before the execution of the current iteration is acquired; the intermediate iteration image is obtained through initial reconstruction or multiple reconstruction of an emission tomography data set; segmenting the intermediate iteration image to obtain a plurality of sub-regions; each sub-region corresponds to different anatomical structures or regions of interest; inputting the at least one sub-region into a trained neural network, and obtaining a pixel prediction value of the sub-region; updating at least one pixel value of the intermediate iteration image based on the pixel prediction value to obtain a corrected intermediate iteration image; and performing activity reconstruction based on the corrected intermediate iteration image to obtain an intermediate iteration image of the current iteration. The method can improve the convergence speed.

A method and a system for imaging a transient event, the method comprising passively recording spatiotemporal projections of the transient event in an angular range from −45° to +45°; and processing the recorded spatiotemporal projections by compressed sensing tomographic image reconstruction to recover the transient event. The system comprises an imaging unit and a shearing unit for imaging a dynamic event to different positions, a detector that records data by spatially integrating over each pixel and temporally integrating; and a processing unit that reconstructs the dynamic event from said data by compressed sensing tomographic image reconstruction.

This invention discloses a method, system, medium, and device for NNBI visual tomographic image reconstruction, relating to the field of neutral beam imaging technology using negative ion sources. The NNBI visual tomographic image reconstruction method and system provided by this invention utilizes a tomographic reconstruction algorithm to invert the observed projection map of an NNBI multi-angle beam image, obtaining an initial reconstructed image and constructing its ROI mask. Then, the mask is input into a pre-trained residual denoising network via two channels to acquire the residual image and calculate the denoised image. This suppresses interference from background and noise outside the ROI and improves the stability of denoising, thereby improving image reconstruction efficiency and image quality. Furthermore, based on a forward projection operator constructed using camera calibration parameters and the observed projection vector, the reprojection residual of the denoised image is calculated and a correction value is generated. This correction value is used to perform data consistency correction on the denoised image to obtain the target reconstructed image, maintaining projection consistency and preventing consistency degradation caused by depth denoising. This invention effectively suppresses isolated points and stripe artifacts during image reconstruction, improving the quality of image reconstruction.

The invention discloses a training method and a synthesis method of a tomography image synthesis model, and the training method comprises the steps: obtaining a training image set which comprises a plurality of training image groups; inputting the training image set into a chordal graph domain conditional diffusion network for conditional diffusion to obtain a trained chordal graph domain conditional diffusion network and a plurality of chordal graph domain images output by the trained chordal graph domain conditional diffusion network; inputting the training image set and all chordal graph domain images into an image domain conditional diffusion network for feature fusion and conditional diffusion to obtain a trained image domain conditional diffusion network; and obtaining a trained tomography image synthesis model according to the trained chordal graph domain conditional diffusion network and the trained image domain conditional diffusion network. According to the training method, a tomography image synthesis model is provided, and the reconstruction and synthesis quality of the tomography image can be improved. The invention relates to the technical field of image processing.

The invention discloses a self-adaptive weighted filtering reconstruction method and device for double-circle trajectory imaging of a radiation source and a detector. The method comprises the following steps: acquiring original projection data; constructing a composite weight matrix by obtaining the dynamic offset of the virtual projection point; generating weighted projection data according to the original projection data and the composite weight matrix; performing periphery filling and image rotation operation on the weighted projection data to obtain an intermediate processing image; performing full-angle self-adaptive frequency spectrum continuous compensation filtering on the intermediate processing image to obtain a compensated image; performing image rotation and intelligent center cutting operation on the compensated image to obtain a final projection; and based on an FDK back projection formula, performing weighted back projection reconstruction on the final projection to obtain a three-dimensional cross-sectional image. The method can realize high-efficiency, high-quality and isotropic-resolution image reconstruction, and can be widely applied to the technical field of three-dimensional cross-sectional image reconstruction.

A time-of-flight positron emission tomography (TOFPET) assembly for detecting lesions of a breast of a subject, wherein the subject may anatomically be defined with a median plane and chest wall-coronal plane. The assembly may comprise: a detector array having at least two or more detector segments. The detector segments may include: a scintillator for placement toward the target, the scintillator having a top edge generally closest to the subject and a detection surface wall aligned closest to surrounding the breast, a photo multiplier opposite the scintillator, and a readout connected to the photo multiplier. The assembly may also comprise a processor that receives the acquired tracer emission signals and converts the signals into a three dimensional, tomographic image reconstruction. The detector array is defined by a ring surrounding the breast and the face of ring that may be tilted to offset the chest wall-coronal plane of the subject, and wherein one of the top edges of one of the detector segments is above the chest wall-coronal plane of the subject in the posterior direction.

The present invention discloses a tomographic reconstruction method and device based on dual-plane long-stroke scanning, which is applied to a dual-plane X-ray machine system. The method comprises the following steps: setting a dual-plane X-ray scanning mode and a geometric environment to obtain original images of an object to be reconstructed in different relative position relationships; preprocessing the original image to obtain grayscale values ​​of the original image, performing logarithmic transformation to convert the grayscale values ​​into corresponding X-ray absorptivity, and determining a tomographic reconstruction range according to the long-stroke scanning geometric environment; selecting a reconstruction width, traversing all reconstruction range points with the reconstruction width, and completing reconstruction of coronal and sagittal tomographic images by filtered back projection; obtaining overlapping parts of the reconstructed tomographic images, and splicing the overlapping parts using a weighted superposition method to form a long-stroke image, thereby realizing a tomographic reconstruction algorithm for dual-plane long-stroke scanning, realizing reconstruction of coronal and sagittal weighted-position tomographic images, expanding the tomographic image reconstruction range, and improving the image inter-layer resolution.

The application discloses a training method and a synthesis method of a tomographic image synthesis model, wherein the training method acquires a training image set, the training image set comprises a plurality of training image groups; the training image set is input into a chord diagram domain conditional diffusion network to perform conditional diffusion, a trained chord diagram domain conditional diffusion network is obtained, and a plurality of chord diagram domain images output by the trained chord diagram domain conditional diffusion network; the training image set and all the chord diagram domain images are input into an image domain conditional diffusion network to perform feature fusion and conditional diffusion, a trained image domain conditional diffusion network is obtained; and a trained tomographic image synthesis model is obtained according to the trained chord diagram domain conditional diffusion network and the trained image domain conditional diffusion network. The training method provides a tomographic image synthesis model, which is beneficial to improving the quality of tomographic image reconstruction and synthesis. The application relates to the technical field of image processing.

The present application relates to a kind of based on extremely sparse scanning computer tomography image reconstruction method and system, comprising: obtaining CT machine original scanning data and the reconstruction image of standard algorithm;Adopt the generation of adversarial network;Design loss function, train the generation of adversarial network;VdCS algorithm and CT-RDNet network are combined through feedback path, construct extremely sparse scanning CT image reconstruction framework, i.e.CSUF image reconstruction framework;Obtain the scanning data of CT machine extremely sparse scanning mode, i.e.extremely sparse scanning data;With extremely sparse scanning data as input, CSUF image reconstruction framework executes extremely sparse scanning CT image reconstruction, and outputs the reconstruction image satisfying the needs of radiology clinical diagnosis.The method of the present application can obtain CT image satisfying the needs of radiology clinical diagnosis under extremely sparse scanning condition, and has good robustness and engineering practicability.