54 results about "Voxel size" patented technology

The invention discloses a voxel map plane merging method, system and device and a storage medium, and the method comprises the steps: S1, carrying out the voxelization of a point cloud, detecting a plane in a voxel, and recording the feature information of the plane; s2, classifying the planes in the voxels according to the plane feature information, and calculating a hash key value; and S3, combining the plurality of planes according to the hash key values and the position relationship of the different planes. According to the method, by calculating the hash key values of the voxel planes, the voxel planes with similar geometric attributes can be quickly merged into the same hash bucket and are uniformly processed, so that the real-time performance and the resource utilization efficiency of the system are greatly improved, and particularly, when a large-scale environment is processed, the real-time performance of the system is greatly improved. The defects of a traditional fixed voxel size method can be more effectively overcome.

The invention discloses a rock mass falling identification algorithm based on point cloud distance tolerance, and belongs to the technical field of rock mass monitoring, and the algorithm comprises the steps: obtaining image data of a collection region, and generating a live-action three-dimensional model and original point cloud data of the collection region; removing the water surface feature points in the point cloud data; performing density adaptive voxel downsampling on the point cloud without the water surface feature points; iCP registration is executed layer by layer by adopting a voxel size progressive decrease strategy, when ICP registration is executed on each layer, iteration convergence is controlled through a dynamic threshold value, and transformation matrixes of all layers are accumulated to obtain a final registration result; and calculating the Hausdorff distance tolerance of the point clouds after the two stages of registration, when the distance tolerance is greater than a preset threshold value, determining the rock mass falling area, and calculating the rock mass falling volume by using an Alpha-Shape algorithm in combination with Delaunay triangulation. The algorithm based on the point cloud distance tolerance is high in recognition precision in the rock mass falling area; the method can be effectively applied to rock mass monitoring and loosening and peeling detection.

The invention provides an improved density-based iterative voxel downsampling method, and more flexible and adaptive point cloud downsampling is realized by introducing point cloud density information and iterative voxel size. According to a traditional voxel filtering method, a fixed voxel size is used, important details can not be effectively reserved in an area with large density change, and excessive downsampling can be caused in a sparse area. According to the method, the point cloud number of each voxel is calculated, and the voxels are dynamically segmented and subdivided according to the preset threshold value, so that more point cloud information is kept in a high-density region, redundant data is reduced in a low-density region, and the downsampling precision and the calculation efficiency are improved. Specifically, when the number of points in a voxel exceeds a set threshold value and the size of the voxel is larger than the minimum size, the algorithm divides the voxel into eight sub-voxels, and recursive subdivision continues until conditions are met. The method is high in adaptability, can dynamically adjust the sampling strategy according to the density change of the point cloud, and remarkably improves the precision and efficiency of point cloud processing. Compared with a traditional voxel filtering method, the method has the advantages that details of a dense region can be better reserved, redundant points of a sparse region are reduced, and an efficient and accurate point cloud downsampling solution is provided.

The invention provides a CT (Computed Tomography) system defect detection capability reference block, evaluation method and system. The CT system defect detection capability reference block comprises a defect simulation sample A and a matrix disc B, the method comprises the following steps: preprocessing a CT detection image, extracting a defect region of interest, and obtaining a defect detection capability evaluation two-dimensional plane; correcting the voxel size of the CT image; carrying out sub-pixel edge detection; fitting a result, and establishing an image pixel position coordinate; according to the sub-pixel edge point set, matrix gray scale mu l (h) and defect gray scale mu b (h) at different positions are obtained; establishing a relational expression of the defect size and the gray level, drawing a relation curve I of the defect size D and # imgabs0 # and a human eye 50% probability recognition probability curve II, and obtaining a minimum recognizable defect; and establishing a relation curve of the defect size and the measurement error thereof. According to the method, the defect size measurement precision is high, the influence of image noise is small, and the minimum defect detection capability and defect measurement error can be quantitatively evaluated.

The invention discloses a maxillary anterior tooth alveolar bone defect identification method and device. The method comprises the following steps: S1, acquiring a global voxel matrix and voxel size information; segmenting the global voxel matrix to obtain N pieces of tooth information; s2, based on the voxel size information, processing each piece of tooth information to obtain a vector set corresponding to each piece of tooth information; s3, processing each piece of tooth information and the corresponding vector set to obtain a classification image pair set corresponding to each piece of tooth information; s4, processing the classification image pair set corresponding to each piece of tooth information by using a pre-trained multi-scale classification recognition model to obtain a probability vector set corresponding to each piece of tooth information; and S5, processing the probability vector set corresponding to each piece of tooth information to obtain a classification identification result corresponding to each piece of tooth information. According to the method, high-precision classification and identification of alveolar bone defect conditions can be automatically completed.

The invention relates to the technical field of machining simulation, in particular to a voxel-based five-axis real-time cutting simulation method, which comprises the following steps of: determining a voxel size according to the number of rendered voxels, and performing voxelization processing on workpiece modeling according to the voxel size to obtain a workpiece voxel grid model; carrying out blocking processing on the workpiece voxel grid model to obtain a plurality of voxel blocks; and updating and rendering the model voxel in the voxel block according to the pose information of the cutter. In order to solve the problem that in the prior art, when large-scale voxels are processed in the cutting simulation process, efficiency is low or precision is insufficient, the size of the rendered voxels is dynamically calculated through workpiece modeling, voxelization processing is carried out, the good rendering efficiency is achieved, block processing is carried out on a workpiece voxel grid model, and the processing efficiency is improved. When cutting is simulated, only the specific block making contact with the cutter is locally updated, the number of voxels needing to be simulated and rendered again is reduced, and the processing efficiency is improved.

Method for monitoring a machining, in particular, subtractive machining of a workpiece (105) with an aid of a CNC machine (104), the method comprising at an interface (101), - receiving (200) input data comprising a motion trajectory data of a tool (106) of the CNC machine (104), wherein, during machining of the workpiece (105), the tool (106) is moving along the motion trajectory, at a computing system (102), - generating (201) a voxel-based representation of the workpiece (105) from the input data, wherein the generating comprises specifying a voxel size so that the voxel size corresponds to a surface roughness of a machined surface of the workpiece, at a displaying device (103), - visualizing (202) the voxel-based representation of the workpiece with a resolution determined by the voxel size.

The invention discloses a method for detecting mouse brain nucleus activation based on manganese-enhanced magnetic resonance imaging, and belongs to the field of image processing, and the method comprises the steps: converting an acquired mouse head manganese-enhanced magnetic resonance image into an NIFTI format, and enabling the direction and voxel size of the image to be consistent with a standard mouse brain map template; performing offset field correction on the image; loading a PLKA-nnUNet model, carrying out image segmentation on the image, outputting a binary brain mask, and extracting an individual mouse brain image from the image after bias field correction; carrying out image registration and intensity normalization; for each registered individual mouse brain image, calculating relaxation rate mean values of four hippocampal subregions of the mouse brain R1 image, wherein the relaxation rate mean values are used for quantitatively comparing mouse brain activation conditions; and carrying out voxel-level statistical test on the registered and normalized individual mouse brain images to identify brain regions with intensity differences among different experimental conditions. According to the method, the dependence on professional operation is reduced.

Method for monitoring a machining, in particular, subtractive machining of a workpiece (105) with an aid of a CNC machine (104), the method comprising at an interface (101), - receiving (200) input data comprising a motion trajectory data of a tool (106) of the CNC machine (104), wherein, during machining of the workpiece (105), the tool (106) is moving along the motion trajectory, at a computing system (102), - generating (201) a voxel-based representation of the workpiece (105) from the input data, wherein the generating comprises specifying a voxel size so that the voxel size corresponds to a surface roughness of a machined surface of the workpiece, at a displaying device (103), - visualizing (202) the voxel-based representation of the workpiece with a resolution determined by the voxel size.

The digital twinning method of the anatomical structure comprises the steps that medical image data of the anatomical structure is analyzed, a three-dimensional array and metadata are obtained, the three-dimensional array comprises density information or gray information of the anatomical structure at different spatial positions, and the metadata comprises voxel sizes; performing live-action rendering on the three-dimensional array to obtain an anatomical picture; generating a scene file based on the metadata, the three-dimensional array and the anatomical picture; and rendering the scene file to obtain the digital twin of the anatomical structure capable of being presented in the set physical space. The digital twinning method of the anatomical structure can generate a high-fidelity digital twinning body of the anatomical structure. The invention further provides a digital twinning system of the anatomical structure.

The application discloses a nasal airflow simulation analysis method based on a labeled voxel calculation domain, an electronic device and a storage medium. The method comprises the following steps: acquiring a three-dimensional surface model of a nasal cavity, and constructing a voxel calculation grid based on a minimum containing box of the three-dimensional surface model and a preset voxel size; performing voxelization processing and label assignment on the voxel calculation grid to automatically construct an initial labeled voxel calculation domain; performing voxel domain processing and automatic marking of an opening boundary on the initial labeled voxel calculation domain to obtain a simulation labeled voxel calculation domain; automatically determining boundary conditions and numerical simulation parameters based on the simulation labeled voxel calculation domain and a preset breathing condition, establishing a numerical simulation model and iteratively solving the numerical simulation model to generate a nasal airflow simulation result. The application realizes full-process automation from a three-dimensional surface model to a airflow simulation result, significantly reduces manual grid division and boundary setting operations, and improves the efficiency and automation degree of nasal airflow simulation.

Embodiments of the present application disclose a mapping method, a computer device and a computer readable storage medium. The method comprises the following steps: obtaining point cloud data of a target detection range, and mapping the point cloud data to corresponding voxels; performing a division operation on the voxels by using a preset fixed truncation distance, so as to divide the voxels into target voxels and interference voxels; filtering out the point cloud data corresponding to the interference voxels; constructing a three-dimensional model of a target object in the target detection range according to the point cloud data corresponding to the target voxels, and outputting the three-dimensional model. Therefore, the present application simplifies the calculation step of the truncation distance, divides the voxels by using the preset fixed truncation distance, reduces the voxel size from the decimeter level to the centimeter level, significantly improves the representation accuracy of indoor small objects, and guarantees the calculation efficiency. Moreover, by using the division of the interference voxels and the target voxels, the misjudgment rate of static points is effectively reduced, and the accuracy and reliability of the dynamic point cloud filtering are improved.

A method for monitoring the machining, in particular the subtractive machining of a workpiece (105) by means of a CNC machine tool (104), comprises: at an interface (101), receiving (200) input data comprising movement trajectory data of a tool (106) of the CNC machine tool (104), in which the tool (106) moves along the movement trajectory during the machining of the workpiece (105); at the computing system (102), a voxel-based representation of the workpiece (105) is generated (201) from the input data, where the generating comprises: specifying a voxel size such that the voxel size corresponds to a surface roughness of a machined surface of the workpiece; at a display device (103),-visualization (202) of the voxel-based representation of the workpiece with a resolution determined by the voxel size.

Described are machine learning platforms for performing three-dimensional volumetric analysis on acoustic neuromas such as vestibular schwannoma. The machine learning platforms contain a machine learning algorithm operably linked to a computer processor. The algorithm is configured to process data originating from medical images from magnetic resonance imaging. Processing the data involves three-dimensional volumetric analysis, utilizing a voxel size of about 0.8x0.8x0.9 mm or smaller, by the algorithm trained using image scans containing raw image scans and validated image scans. The data show that platforms containing a machine learning algorithm described herein not only speed up the segmentation process (less than 4 minutes per scan) but also offer improved depictions and measurements of tumor growth. This can enhance clinical decision-making, provide consistent monitoring of diseased tissue progression / regression (e.g., tumor progression / regression), and empower patients with detailed insights into their condition, thus facilitating better self-advocacy and treatment planning.

Voxel-Interface 3D Printing (VI3DP) enables comprehensive control over extruded voxel interfaces irrespective of the printhead diameter that conventionally dictates feature size. The present invention includes various optical, mechanical, and electrical functionalizations, attaining interface thicknesses up to three orders of magnitude smaller than the voxel size. Notable applications include encoding data in soft matter through fluorescent interfaces, creating tight fits and movable mechanisms through non-adhesive interfaces, fabricating bio-inspired composites with tailored failure modes, and developing a single filament capacitive touch sensor. VI3DP opens new avenues for enhanced functionality and efficiency across multiple fields, including biomedical technology, electronics, optics, and nanotechnology.

A data processing method includes: receiving a plurality of first and second scan data corresponding to a preset environment, the first and second scan data corresponding to first and second position ranges in the preset environment, respectively; thinning the first scanning data and the second scanning data according to a relation graph to generate a plurality of first thinning data and a plurality of second thinning data; uniformly sampling the first thinning data and the second thinning data according to the first voxel size to generate a plurality of first sampling data and a plurality of second sampling data; generating a plurality of smoothing data according to the first sampling data and the second sampling data; dividing the smoothed data into a plurality of data clusters; optimizing the relative relationship of the data clusters and obtaining a state matrix; optimizing a relative relationship of a plurality of nodes of each data cluster according to the state matrix; and generating point cloud map information according to the data cluster.

The invention discloses a surveying and mapping system for a bridge arch rib line shape, relates to the technical field of bridge surveying and mapping, and solves the technical problems that an original point cloud screening and clustering method is poor in adaptability and key feature points are easy to be excessively simplified. In combination with targeted layout of high-reflectivity reflectors and a blind area complementary scanning mechanism, the point cloud coverage rate of the arch rib is improved, scanning blind areas of the inner side and a node area are eliminated, point cloud splicing errors are reduced, the coverage rate of a traditional experience layout scheme is improved, and the coverage rate of a material adaptive reflection intensity threshold value, a dynamic K value and multi-condition constraint clustering is improved. The main point set purity and the background noise rejection rate are improved, the main point recognition accuracy is improved, noise residues and main point loss are avoided, the fragment length and the voxel size are dynamically adjusted based on the curvature and the point density, feature point correlation points are preferentially reserved, the data size is reduced, and the key feature retention rate is improved.

The invention discloses a voxel map plane merging method, system and device and a storage medium, and the method comprises the steps: S1, carrying out the voxelization of a point cloud, detecting a plane in a plurality of voxels, and recording the feature information of the plane; wherein the plane is a flat surface with a certain area; s2, classifying the planes in the voxels according to the plane feature information, and calculating a hash key value; and S3, according to the position relationship between the hash key values and the planes, merging the plurality of adjacent planes. According to the method, by calculating the hash key values of the voxel planes, the voxel planes with similar geometric attributes can be quickly merged into the same hash bucket and are uniformly processed, so that the real-time performance and the resource utilization efficiency of the system are greatly improved, and particularly, when a large-scale environment is processed, the real-time performance of the system is greatly improved. The defects of a traditional fixed voxel size method can be more effectively overcome.

The invention relates to the technical field of three-dimensional modeling, in particular to an oral cavity three-dimensional scanning and data reconstruction method based on laser diffraction, and the method comprises the steps: carrying out the neighbor search of an initial oral cavity point cloud model, and obtaining a curvature ratio change curve of each search radius; determining a target curvature threshold value according to the fluctuation condition of each curvature proportion change curve; performing grading processing by using the target curvature threshold to obtain each curvature grade; obtaining an updated point cloud area of each first curvature level, each point cloud area corresponding to each matched curvature level and a point cloud area of each residual curvature level, analyzing the space size and level of the point cloud areas to determine the size of a thinning voxel, and carrying out thinning processing on the point cloud areas of the corresponding curvature levels. And obtaining a diluted oral cavity point cloud model. According to the method, the key oral tissue characteristics are effectively reserved while the thinning effect of the oral point cloud model is ensured, and data reconstruction is facilitated.

The embodiment of the invention discloses a mapping method, computer equipment and a computer readable storage medium. The method comprises the following steps: obtaining point cloud data of a target detection range, and mapping the point cloud data to corresponding voxels; performing division operation on the voxels by using a preset fixed cutoff distance so as to divide the voxels into target voxels and interference voxels; filtering the point cloud data corresponding to the interference voxels; and constructing a stereoscopic model of the target object in the target detection range according to the point cloud data corresponding to the target voxel, and outputting the stereoscopic model. Therefore, the calculation step of the cutoff distance is simplified, the voxel is divided through the preset fixed cutoff distance, the size of the voxel is reduced to the centimeter level from the decimeter level, the expression precision of the indoor small object is remarkably improved, and meanwhile the calculation efficiency is guaranteed. And the interference voxels and the target voxels are divided, so that the misjudgment rate of static points is effectively reduced, and the accuracy and reliability of dynamic point cloud filtering are improved.

The present application relates to the technical field of engineering safety control, in particular to a three-dimensional real-time map construction method for power transmission and transformation engineering, which sequentially collects substation site clouds, determines initial voxel size according to target volume of drawings, creates voxel grid for thinning, retains any point in voxel space, constructs a three-dimensional map model in sequence according to regions, periodically detects the completeness of model features, if the accuracy is unqualified, determines the reason according to the variance of the difference between target size and drawings, and corrects the thinning mode, thinning parameters and data compression rate; if it is qualified, a three-dimensional real-time map is generated, the target live state, voltage level and safety distance are updated, and a safety operation virtual fence is constructed. The construction method can accurately thin the point cloud data according to the state of the point cloud data and the actual operation demand, thereby efficiently constructing a high-precision three-dimensional map, and providing more accurate and reliable safety protection for nearby live operation.

The invention relates to the technical field of point cloud data processing, in particular to a BIM model construction method considering initial geometric defects of a steel structure, and the method comprises the steps: obtaining point cloud data of each site of the steel structure; dividing the point cloud data of each site into a plurality of voxels; obtaining a first characteristic value of each point according to the distance between each point and the adjacent point; acquiring a second characteristic value of each point according to the coordinate dispersion degree of each point and the adjacent point, and acquiring a characteristic vector of each point in combination with the curvature of each point and the first characteristic value; and according to the first feature values and the second feature values of all the points in each voxel and the distances between all the points and the feature vectors of the adjacent points, obtaining an adjustment coefficient of each voxel, further obtaining a small voxel side length corresponding to each voxel, segmenting each voxel into a plurality of small voxels, further obtaining filtered point cloud data, and constructing a BIM model of the steel structure. According to the method, the sizes of the voxels at different positions are adaptively calculated, so that the construction precision of the BIM model is improved.

In order to solve the problem of low data management efficiency in the prior art, the invention provides a three-dimensional geological information management method based on a GIS (Geographic Information System) and voxels. The method comprises the following steps: collecting geological exploration data; s2, establishing a GIS three-dimensional geological data information model by using the geological exploration data collected in the step S1; s2, establishing a GIS geological boundary and three-dimensional voxel information model by utilizing the GIS three-dimensional geological data information model established in the step S2; s3, establishing a CityGML model of geological information by utilizing the GIS geological boundary and the three-dimensional voxel information model established in the step S3; and S4, according to the CityGML model of the geological information established in the step S4, developing a visual interface. According to the three-dimensional geological information management method based on the GIS and the voxel, the management efficiency and the flow efficiency of the geological information can be improved from a three-dimensional level, and three-dimensional visualization can be directly carried out on the geological information from several levels of the voxel, the size and the attribute.

The present invention provides a method and device for calculating the loading and unloading rate of a carriage. The method includes: obtaining the point cloud data inside the carriage and the size information of the carriage; dividing the point cloud data into multiple voxels according to the preset voxel size information; determining the type of each voxel according to whether the voxel includes valid point cloud and the position of the voxel in the depth direction; calculating the total volume of the multiple voxels according to the preset volume calculation method corresponding to the type of the voxel; and calculating the loading and unloading rate of the carriage according to the total volume and the size information of the carriage. The present invention can divide the point cloud data of the carriage into multiple voxels, classify each voxel according to different cargo stacking methods, and preset different volume calculation methods for different types of voxels. After obtaining the total volume, the loading rate of the carriage can be calculated, which can improve the operation efficiency and the accuracy of the loading and unloading rate.

Systems and methods for digitally simulating a physical process in a three-dimensional CAD model of a simulation space include receiving a digital representation of a simulation space, the digital representation including a three-dimensional CAD model of the simulation space including a lattice structure represented as a plurality of voxels including particles, the simulation space including one or more surfaces sized and oriented independently of the voxels. A plurality of planar facets are determined to represent the one or more surfaces of the simulation space in the lattice structure. One or more curved facets are determined based on the plurality of planar facets; and a physical process is simulated by performing surface interactions between the one or more curved facets and the particles in one or more voxels adjacent the one or more curved facets.

The present application relates to a kind of based on limiting hierarchical voxel structure numerical control machining three-dimensional geometry simulation method, belong to numerical control machining simulation technical field, solve the problem of reduced efficiency in the prior art in finding voxel, simulation memory consumption significantly rises and simulation efficiency reduces, including S1: voxel hierarchical structure initialization, based on the three-dimensional geometry data of the blank model to be processed obtained, establish primary voxel, use as the root node of voxel hierarchical structure;S2: determine final stage voxel size, with the voxel between primary voxel and final stage voxel as intermediate level voxel, determine the subdivision granularity of each hierarchical voxel;S3: based on the intersection detection of voxel of tool motion, determine the intermediate level voxel acted by tool at current simulation time;S4: final stage voxel dynamic generation, according to the spatial size of final stage voxel, generate final stage voxel for geometry simulation calculation;S5: based on the geometry simulation calculation of final stage voxel, obtain the three-dimensional geometry simulation result of numerical control machining process.

The invention discloses a FAST site dangerous rock identification method based on unmanned aerial vehicle intelligent routing inspection, which comprises the following steps: S1, performing point cloud downsampling by using unmanned aerial vehicle routing inspection, obtaining rock wall radar data near a FAST site, and obtaining dangerous rock point cloud data; s2, determining the voxel size and the voxel center of the dangerous rock point cloud data; s3, extracting the features of the dangerous rock target, and extracting the feature vector of each voxel as the shape information of the dangerous rock target; and S4, candidate dangerous rock target detection: extracting a normal vector of the point cloud as a feature vector to describe shape information of the point cloud, and scanning point cloud data by using a sliding window method to obtain all possible candidate dangerous rock bodies. The method adopts intelligent identification and automatic determination, is an efficient and convenient dangerous rock identification method, is safe and efficient, is controllable in cost, has important engineering application value, and has important significance for stable and reliable operation and scientific achievement output of a 500-meter-caliber spherical radio telescope FAST of a national important scientific and technological infrastructure.

The application discloses a method for detecting mouse brain nucleus activation based on manganese-enhanced magnetic resonance imaging, and belongs to the field of image processing, which comprises the following steps: converting the manganese-enhanced magnetic resonance image of the mouse head collected into NIFTI format, keeping the direction and voxel size of the image consistent with the standard mouse brain atlas template; performing bias field correction on the image; loading the PLKA- nnUNet model, performing image segmentation on the image, outputting a binary brain mask, and extracting individual mouse brain images from the image after bias field correction; performing image registration and intensity normalization; calculating the mean relaxation rate of the four hippocampal subregions of the mouse brain R1 map for each individual mouse brain image after registration, which is used for quantitatively comparing the activation of the mouse brain; and performing statistical testing at the voxel level on the registered and normalized individual mouse brain images to identify the brain regions with intensity differences between different experimental conditions. The method reduces the dependence on professional operation.

The invention provides a dose distribution determination method, a dose distribution determination device and electronic equipment, which can be applied to the technical field of medical radiation dose calculation. The method comprises the steps of performing pixel conversion on a cross-sectional image based on tissue density data related to a target animal to obtain a three-dimensional density matrix; generating a primary simulation particle beam by using a normal distribution sampling algorithm based on a preset energy acquisition density matched with the voxel size of the cross-sectional image; the energy range and the beam shape of the primary simulation particle beam are adjusted on the basis of preset phase space parameters determined according to the tumor shape of a target area of the target animal, an adaptive particle beam is obtained, and the tumor shape of the target area is determined on the basis of the cross-sectional image; in the process of simulating transportation of the adaptive particle beam in the tissue in the target area represented by the three-dimensional density matrix, the Poisson distribution model is used for processing energy change information generated when the adaptive particle beam generates the simulated physical effect in the tissue, and a dose distribution result is obtained.

The embodiment of the invention provides a point cloud data processing method and device, equipment, a storage medium and a program product, and relates to the field of big data. The method comprises the steps of obtaining three-dimensional point cloud data of a target monitoring object, processing a point cloud space corresponding to the three-dimensional point cloud data, obtaining a neighbor point set of each data point to determine a surface change feature value of each data point, and determining the data points with the surface change feature values larger than a preset edge threshold value as edge feature points. Determining an initial clustering center based on the non-edge feature points and a preset voxel size, and performing clustering processing on each non-edge feature point and the current clustering center to obtain a plurality of clusters; and determining the number of data points needing to be reserved in each clustering cluster according to a preset simplification rate, so as to extract clustering feature points. And carrying out merging processing on the edge feature points and the clustering feature points to generate a simplified point cloud data set. According to the method, the accuracy and the monitoring efficiency of the bank loan project monitoring result are improved.