9results about How to "Guaranteed registration accuracy" patented technology

The invention discloses a point cloud registration method and device, electronic equipment and a storage medium, and the method comprises the steps: firstly obtaining an image collected by image collection equipment and point cloud data synchronously collected by a laser radar; the image and the point cloud data are preprocessed to obtain a corrected image and a corrected point cloud, so that unnecessary interference of other data on point cloud registration is avoided; when registration is carried out, coarse registration is carried out on the corrected point clouds based on each frame of image in the corrected image and the corresponding corrected point clouds to obtain coarse registration point clouds, so that the registration precision of the coarse registration is ensured; after coarse registration is completed, a search radius is set, fine registration is performed on the coarse registration point cloud based on the search radius, and a target point cloud corresponding to each frame of image in the corrected image is obtained, so that the point clouds of dynamic objects at different positions in adjacent frames can be effectively proposed, and the efficiency of searching for corresponding points can be improved; therefore, the precision and efficiency of registration are ensured.

The invention discloses a cultivated land non-agrochemical intelligent monitoring method and system based on multi-temporal SAR and optical image fusion, and belongs to the technical field of remote sensing monitoring and land resource management. The method solves the problems of information loss and misjudgment caused by cloud and mist shielding or noise interference of a single remote sensing data source. According to the scheme, the method comprises the following steps: acquiring and preprocessing a multi-temporal SAR and an optical image sequence; carrying out space registration; constructing a double-branch deep neural network, respectively extracting radar scattering features and spectral texture features, and fusing the radar scattering features and the spectral texture features by adopting a cross attention mechanism; generating a change intensity graph based on a time sequence change detection module; and extracting non-agrochemical suspected pattern spots through threshold segmentation, and aggregating and outputting a final monitoring result. The method is mainly used for realizing the automatic and high-precision monitoring of the non-agrochemical change of the cultivated land.

The invention provides a method for improving earth observation precision of atmospheric detection loads. The method comprises the following steps: synchronously correcting different atmospheric detection loads; synchronously correcting multiple detection channels of the atmospheric detection load; and synchronously correcting the atmospheric detection load observation angle, the data acquisition time and various satellite parameters. The method has the beneficial effects that a method of synchronizing a plurality of physical signals and eliminating software monitoring errors is adopted for observing and scanning the same observation target by different atmospheric detection loads of a commercial satellite, so that time observation errors caused by time delay and data receiving of different detection loads of a traditional satellite through a CAN (Controller Area Network) bus or 1553 bus and time broadcast are eliminated; observation errors between different detection channels and different acquisition points are eliminated, so that the accuracy of acquiring the same observation target data by different detection loads is improved.

The invention provides an image registration method, system and device and a medium. The image registration method comprises the steps of obtaining a to-be-registered image and a reference image of a target object; wherein at least one of the to-be-registered image and the reference image is a Histoimage of a PET (positron emission tomography) mode; and carrying out image segmentation processing on the to-be-registered image and the reference image, carrying out image registration processing on the to-be-registered image and the reference image after segmentation processing, and generating a deformation field for registering the to-be-registered image to the reference image so as to realize image registration of the to-be-registered image. According to the method and the device, the Histoimage with good structural information is quickly generated in real time in the PET scanning process, the Histoimage is used for replacing the reconstructed image in the prior art, the registration precision of the PET modal image is ensured, the real-time performance in image registration is improved, and the calculation overhead is reduced.

The application discloses a parameterized human body model reconstruction method based on a Chinese database, three-dimensional human body grid data of multiple female naked bodies of different ages, body types and heights are collected to establish a database; point clouds are learned based on BPS, SMPL model grid points are respectively roughly registered to the vicinity of each three-dimensional human body grid point to generate corresponding rough registration models; then, a non-rigid grid registration algorithm based on an ICP algorithm is used to finely register each rough registration model to generate corresponding SMPLD models; each SMPLD model is corrected in a standard standing posture, and then PCA principal component analysis is performed on each SMPLD model after the standing posture correction, a plurality of principal component proportions obtained are used as body parameters of a parameterized human body model adapted to Chinese body shapes, original posture parameters of the SMPL model are used as posture parameters of the parameterized human body model adapted to Chinese body shapes, and a final parameterized human body model is constructed.

This invention provides a GAN-based method for motion compensation in prostate biopsy via TRUS, comprising: acquiring intraoperative 2D TRUS images and preoperative 3D TRUS images; inputting the intraoperative 2D TRUS images into a pre-trained registration network, which is trained as a generator in a generative adversarial network; the registration network predicting and outputting spatial transformation parameters between the intraoperative 2D TRUS images and the preoperative 3D TRUS images based on the features of the intraoperative 2D TRUS images; and applying the spatial transformation parameters to the preoperative 3D TRUS images to achieve motion compensation of the prostate. This invention integrates a Self-Attention Feature Reassignment (SAFR) module to achieve accurate extraction of the overall structure and boundaries of the prostate, and designs a physical decoupling regression head at the output of the registration network to achieve separation of physical dimensions, gradient balancing, and output range control.

The application discloses a three-dimensional image registration method, comprising the steps of: obtaining the position of a mark point on a phantom in a phantom coordinate system; obtaining a three-dimensional image of the phantom through a C-arm scanner, periodically obtaining the pose of the C-arm in the scanning process, and constructing a C-arm coordinate system; identifying the position of the mark point in the three-dimensional image, and calculating the transformation relationship between the image coordinate system and the phantom coordinate system; calculating the transformation relationship between the C-arm coordinate system and the phantom coordinate system according to the pose of the C-arm and the phantom identified by an optical tracking system, and calculating the transformation relationship between the C-arm coordinate system and the image coordinate system. In the application, the use of intraoperative positioning mark points is removed, the registration accuracy is ensured, the surgical process is simplified, and the surgical efficiency is improved.

The invention discloses a chip image registration method and device, equipment and a storage medium, and relates to the technical field of image processing, and the chip image registration method comprises the steps: obtaining a chip image which comprises first resolution image information; performing down-sampling on the chip image to obtain second resolution image information corresponding to the chip image; the second resolution corresponding to the second resolution image information is smaller than the first resolution corresponding to the first resolution image information; performing first registration verification on the chip image according to the second resolution image information to obtain a first verification value, and responding to the condition that the first verification value is greater than a first verification threshold to obtain target registration information; or, in response to the fact that the first verification value is smaller than the first verification threshold value, second registration verification is conducted on the chip image according to the first resolution image information, and target registration information is obtained. Through the above mode, registration is carried out in combination with images with different resolutions, and on the premise of ensuring the registration precision, the registration operand is reduced, and the registration efficiency is improved.

The application discloses a cartoon image toy 3D scanning reverse modeling copying processing method, which comprises the following steps: selecting a cartoon image toy prototype to be copied, removing detachable parts, and spraying a matt auxiliary agent on the detail feature area of the cartoon image prototype; scanning the cartoon image toy prototype, and specially scanning the detachable parts and the detail feature area to obtain three-dimensional point cloud data; performing trajectory fitting processing on the edge extension position of the detachable parts; and performing weighted data splicing on the three-dimensional point cloud data. The application adopts differential high-precision scanning and layered progressive scanning, accurately captures complex structures such as detail feature areas, restores the arc, gap and limiting features, performs trajectory fitting and curvature continuity processing on the edge extension section of the detachable parts, solves the problems of misconnection, too large gap and poor cooperation, and greatly improves the assembly precision and smoothness.