16 results about "Spatial change" patented technology

The embodiment of the invention provides a three-dimensional space change detection method and device, equipment and a storage medium, and relates to the field of digital image processing. The method comprises the following steps: acquiring image sequences of the same area shot by the unmanned aerial vehicle in two periods; for each image sequence, performing point cloud reconstruction on the region based on the image sequence to obtain the group of point cloud data; point cloud registration is carried out according to the similarity between the feature descriptors of all the points in the two sets of point cloud data, a matching point pair set comprising a plurality of matching point pairs is obtained, and two points in each matching point pair belong to two pieces of point cloud data and are closest to each other; and for each matching point pair, geometric information of two points in the matching point pair in respective point cloud data is acquired, and if it is determined that the difference between the two pieces of geometric information meets a preset difference condition, it is determined that the corresponding position of the matching point pair in the region is changed. According to the embodiment of the invention, the problem of insufficient spatial change detection precision in a complex scene is solved.

The invention relates to the technical field of remote sensing images, in particular to a remote sensing image semantic change detection method based on multi-head adaptive attention and spatial feature enhancement, based on an MAA-SFENet model architecture, a multi-head adaptive attention (MAA) and attention driven feature association (ADFC) module is integrated, the MAA focuses on capturing fine local changes, and the ADFC module focuses on capturing fine local changes. And meanwhile, the global context is fused, and the ADFC focuses on processing large-scale high-dimensional data, so that the sensitivity to a spatial variation mode is enhanced. Besides, in order to further improve the adaptability to different environments, a spatial adaptive feature modulation (SAFM) module is introduced, and spatial information representation is dynamically adjusted, so that the detection precision under a complex background is improved. Experimental results on two SCD data sets show that the MAA-SFENet is superior to other advanced methods in qualitative and quantitative evaluation. The ablation experiment further verifies the important contribution of the three modules to performance improvement. Experimental results show that the MAA-SFENet has excellent performance and reliability in the remote sensing image SCD.

An apparatus, method and computer program is described comprising: detecting, at a second user device, one or more display patterns displayed at a display of a first user device, wherein the one or more display patterns comprise features with spatial variations configured to be detected by a radar module of the second user device; determining at least one of an orientation, position, or size of the first user device based, at least in part, on the one or more display patterns; and determining at least one of an orientation or position of the second user device relative to the first user device within a coordinate system of the first user device.

In seismic data, migration images and inverted models, interesting structures with sharp spatial variations such as faults, channels or caves are embedded in laterally smooth stratigraphic layers, coherent and non-coherent noises etc. One way to enhance these structures is to take advantage of the local strong contrast property and use Sobel filter. However, Sobel filter is a discrete differentiation operator which enhances the edges and may lead to drastic amplitude changes within one feature and add difficulties to interpretation. The method uses histogram equalization after Sobel filter which makes the amplitude of a feature much more homogenous and presents a much clearer image of faults or channels with little background noise.

Method for locating a tool (2) in an X-ray image, wherein the tool (2) has an X-ray absorption strength (12) that is spatially variable along a predetermined longitudinal direction of the tool (2) according to a predetermined absorption characteristic, and wherein - a spatial frequency spectrum of a partial area (6) of the X-ray image is determined; - checking whether the spatial frequency spectrum corresponds to the absorption characteristic of the tool (2); and - if the inspection reveals that the spatial frequency spectrum corresponds to the absorption characteristic of the tool (2), the partial area (6) of the X-ray image is determined as the position of the tool (2), whereby - according to the absorption characteristic, the spatial change in the X-ray absorption strength (12) along the longitudinal direction has a characteristic frequency at which a reference spatial frequency spectrum (13) of the spatial change in the X-ray absorption strength (12) along the longitudinal direction has a peak value; and - to check whether the spatial frequency spectrum corresponds to the absorption characteristic of the tool (2), it is checked whether the spatial frequency spectrum has a peak value at a point which corresponds to the characteristic frequency within a predetermined tolerance range.

The invention relates to the technical field of human-computer interaction, and discloses a motion track reproduction method and device, computer equipment and a storage medium. The method comprises the steps of collecting respective motion data of a first object and a second object; based on the motion data, acquiring spatial change information of the first object relative to the second object, including relative pose change and relative position change of the first object; spatial change information of the first object relative to the second object is analyzed, and independent motion data, including pose change and position change, of the first object in the space relative to the second object is separated; and reproducing a motion track of the first object based on the independent motion data. The motion data of the first object and the motion data of the second object are subjected to conjoint analysis, the relative space relation between the first object and the second object is analyzed, interference factors generated by driving of the second object are effectively eliminated, the independent motion data of the first object are extracted, and misjudgment caused by coupling influence in a traditional method is avoided.

The utility model discloses a three-dimensional holographic detection cradle head, which belongs to the inspection robot technology, solves the problem that the three-dimensional holographic space variation detection of a target object cannot be realized in the prior art, and adopts the technical scheme that the three-dimensional holographic detection cradle head mainly comprises a first bin and a second bin, the second bin is provided with a 3D laser module, a high-definition camera module and / or an infrared module, and the 3D laser module, the high-definition camera module and the infrared module can be in signal connection with the robot. The horizontal driver is used for driving the first bin and the second bin to rotate horizontally, the pitching driver is used for driving the second bin to rotate in a pitching mode, and the horizontal driver can lock the first bin and the second bin at a target horizontal angle. The pitch driver is capable of locking the second bin at a target pitch angle. The three-dimensional holographic space change detection device is mainly used for detecting the three-dimensional holographic space change of the target object.

An embodiment relates to a computer-implemented method for denoising a noisy image. The method includes obtaining a noisy input image and estimating a time map (214) based on the noisy input image and noise in the noisy input image, where the time map is spatially varying, and the method includes performing a diffusion process on the noisy input image over two or more time steps. At each time step, the method includes estimating a denoised image based on the noisy input image, a current time-step image, and a time step using a machine-learned model; obtaining a next time-step image based on the denoised image and the current time-step image; updating, for each pixel, the next time-step image with a corresponding value from the denoised image; setting the next time-step image as the current time-step image; and decrementing the time step. The estimating, obtaining, updating, setting, and decrementing are repeated while the time step is positive. The method further includes outputting the denoised image.

A method for modifying an upper layer of a workpiece (152, 252) using a gas cluster ion beam, GCIB, comprising the following steps: Acquiring parameter data relating to an upper layer of a workpiece (152, 252), the parameter data comprising an initial profile of a measured attribute of the upper layer and at least one n-th order derivative of a spatial change of the measured attribute, where n is an integer greater than zero; determining a predicted systematic error response for applying a GCIB (128, 128A) to the upper layer to change the initial profile of the measured attribute using the parameter data comprising the at least one n-th order derivative as input to a predetermined predictive model; Identifying a target profile of the measured attribute; Directing a gas cluster ion beam, GCIB (128, 128A), to the upper layer of the workpiece (152, 252) and spatially modulating an applied property of the GCIB (128, 128A) based, at least in part, on the predicted systematic error response and the parameter data as a function of position on the upper layer of the workpiece (152, 252) to achieve the target profile of the measured attribute.

This application relates to the field of optical measurement technology, and in particular to a thickness detection method and a non-contact optical scanning system, comprising: acquiring a two-dimensional characterization image of the sample to be measured; determining an effective detection area; planning a detection path; acquiring raw thickness data of multiple feature sampling points along the detection path; identifying candidate anomalies among the feature sampling points, wherein candidate anomalies are feature sampling points whose spatial change rate exceeds a set threshold; after identifying candidate anomalies, a first correction process can be executed, including: performing region segmentation on the two-dimensional characterization image to generate an attribute mask for distinguishing regions with different physical properties; extracting local neighborhood attribute features of the corresponding positions of candidate anomalies in the attribute mask; correcting the attributes of candidate anomalies; and filtering the raw thickness data of multiple feature sampling points based on the corrected attributes to generate the thickness of the sample to be measured. This application can demonstrate extremely high efficiency in high-precision measurement and large-scale automated production.

An apparatus, method and computer program is described including: generating a display pattern including features with spatial variations, wherein the spatial variations are configured to enable determination, by a first radar device, of at least one of an orientation of a first user device, position of the first user device, size of a display of the first user device, or one or more gestures associated with the display.

The invention discloses a small spatial variation aberration optical design method, which comprises the following steps of: determining a performance index and an initial structure parameter of an optical system according to an imaging requirement of a target imaging system; based on the initial structure parameters, calculating vertical height variances and horizontal width variances of point spread functions under different view field sampling points, peak signal-to-noise ratios and structural similarity indexes of view field point sequence diagrams and center view field point sequence diagrams, and peak signal-to-noise ratios and structural similarity differences of the point sequence diagrams caused by defocus; the performance of the optical system under the defocusing condition is optimized; and calculating a comprehensive evaluation function used for evaluating the spatial aberration change condition of the optical system, iteratively optimizing lens structure parameters through a least square method, and when iteration is finished, outputting a final optimized optical system structure. According to the invention, by introducing the comprehensive evaluation function for carrying out aberration evaluation on different view fields and out-of-focus distances, optimization design is carried out on the optical system, and the uniformity and the stability of the spatial aberration of the optical system are remarkably improved.

The invention relates to the technical field of image processing, and particularly discloses a coordinate mapping method and device for interaction points of a three-dimensional object, equipment and a medium. At least one two-dimensional layer of a three-dimensional object is obtained, and a first target layer in a three-dimensional space is obtained according to the two-dimensional layer and a conversion matrix, so that intersection of rays and the layers occurs in the three-dimensional space, coordinate influence caused by space change when a to-be-mapped interaction point is mapped from the three-dimensional space to a two-dimensional space is avoided, and the mapping accuracy is improved. Ensuring that the geometrical relationship of each to-be-mapped interaction point on the three-dimensional object before and after mapping is unchanged; when the ray and the first target layer have the intersection point, obtaining the relative coordinate of the intersection point based on the intersection result, and combining the relative coordinate of the intersection point with the layer coordinate of the intersection layer to obtain the mapping coordinate of the to-be-mapped interaction point, thereby avoiding the coordinate influence caused by the spatial change of the layer; and the geometric relationship of the to-be-mapped interaction points before and after mapping is further ensured to be unchanged, so that the mapping accuracy of the interaction points in the three-dimensional object is improved.

A method for determining dimensional data of an object includes emitting terahertz radiation by a first transmitter onto a surface of the object. The terahertz radiation emitted by the first transmitter is received by a first receiver after the terahertz radiation has passed through the object. Terahertz radiation comprising a bandwidth of less than 5% of a carrier frequency of the terahertz radiation is emitted by a second transmitter onto at least one location on the surface of the object. The terahertz radiation emitted by the second transmitter is received by a second receiver after the terahertz radiation has passed through the object. A dimension of the object is determined from one of: (i) the terahertz radiation received by the second transmitter; and (ii) a temporal or spatial change of the terahertz radiation received by the second receiver in consideration of that received by the first receiver.

Disclosed in the present invention are a spatial change data linked update method, system and device, and a storage medium. The method comprises: when it is detected that a data change occurs in an original geographic data set, determining a candidate data set spatial grid code corresponding to data set information to be updated; on the basis of the candidate data set spatial grid code and an incremental package spatial grid code, determining associated data set information; determining a unique code corresponding to update data in an incremental package, and determining associated data in the associated data set information on the basis of the unique code; and on the basis of the update data, performing a linked update on the data set information to be updated and the associated data by means of an update rule. In the present invention, by means of the uniqueness, association and positioning characteristics of a unique code, specific data needing to be updated is quickly identified, thereby improving the update processing efficiency.

The invention provides an all-weather robust multimode image fusion method driven by a fine-grained language instruction. The method comprises the following steps: firstly, analyzing a language instruction input by a user to obtain a function condition of a degradation type which is specified to be removed, and determining a space condition of a region which needs to be enhanced; and then processing function conditions and space conditions through a multi-condition coupling network to generate composite control priori, thereby providing a high-quality interaction medium for a dynamic adjustment fusion process. And further seamlessly embedding the generated compound control prior into the fusion process by using a mixed attention fusion module, thereby realizing controllable information fusion. And finally, under the joint constraint of degradation perception reconstruction loss and language feature alignment loss, driving the network to follow a user instruction, and realizing robust fusion in a composite degradation scene with spatial variation. The method has better degradation removal capability and information integration capability, realizes high-quality scene representation under severe conditions, and can provide clean, clear and pleasant fusion visual results.