Background Three-dimensional(3D)shape representation using mesh data is essential in various applications,such as virtual reality and simulation technologies.Current methods for extracting features from mesh edges or ...Background Three-dimensional(3D)shape representation using mesh data is essential in various applications,such as virtual reality and simulation technologies.Current methods for extracting features from mesh edges or faces struggle with complex 3D models because edge-based approaches miss global contexts and face-based methods overlook variations in adjacent areas,which affects the overall precision.To address these issues,we propose the Feature Discrimination and Context Propagation Network(FDCPNet),which is a novel approach that synergistically integrates local and global features in mesh datasets.Methods FDCPNet is composed of two modules:(1)the Feature Discrimination Module,which employs an attention mechanism to enhance the identification of key local features,and(2)the Context Propagation Module,which enriches key local features by integrating global contextual information,thereby facilitating a more detailed and comprehensive representation of crucial areas within the mesh model.Results Experiments on popular datasets validated the effectiveness of FDCPNet,showing an improvement in the classification accuracy over the baseline MeshNet.Furthermore,even with reduced mesh face numbers and limited training data,FDCPNet achieved promising results,demonstrating its robustness in scenarios of variable complexity.展开更多
To estimate basal water storage beneath the Antarctic ice sheet, it is essential to have data on the three-dimensional characteristics of subglacial lakes. We present a method to estimate the water depth and surface a...To estimate basal water storage beneath the Antarctic ice sheet, it is essential to have data on the three-dimensional characteristics of subglacial lakes. We present a method to estimate the water depth and surface area of Antarctic subglacial lakes from the inversion of hydraulic potential method. Lake Vostok is chosen as a case study because of the diverse and comprehensive measurements that have been obtained over and around the lake. The average depth of Lake Vostok is around 345±4 m. We estimated the surface area of Lake Vostok beneath the ice sheet to be about 13300±594 km^2. The lake consists of two sub-basins separated by a ridge at water depths of about 200–300 m. The surface area of the northern sub-basin is estimated to be about half of that of the southern basin. The maximum depths of the northern and southern sub-basins are estimated to be about 450 and 850 m, respectively. Total water volume is estimated to be about 4658±204 km^3. These estimates are compared with previous estimates obtained from seismic data and inversion of aerogravity data. In general, our estimates are closer to those obtained from the inversion of aerogravity data than those from seismic data, indicating the applicability of our method to the estimation of water depths of other subglacial lakes.展开更多
Dithering optimization techniques can be divided into the phase-optimized technique and the intensity-optimized technique. The problem with the former is the poor sensitivity to various defocusing amounts, and the pro...Dithering optimization techniques can be divided into the phase-optimized technique and the intensity-optimized technique. The problem with the former is the poor sensitivity to various defocusing amounts, and the problem with the latter is that it cannot enhance phase quality directly nor efficiently. In this paper, we present a multi-objective optimization framework for three-dimensional(3D) measurement by utilizing binary defocusing technique. Moreover, a binary patch optimization technique is used to solve the time-consuming issue of genetic algorithm. It is demonstrated that the presented technique consistently obtains significant phase performance improvement under various defocusing amounts.展开更多
The image shape feature can be described by the image Zernike moments. In this paper, we points out the problem that the high dimension image Zernike moments shape feature vector can describe more detail of the origin...The image shape feature can be described by the image Zernike moments. In this paper, we points out the problem that the high dimension image Zernike moments shape feature vector can describe more detail of the original image but has too many elements making trouble for the next image analysis phases. Then the low dimension image Zernike moments shape feature vector should be improved and optimized to describe more detail of the original image. So the optimization algorithm based on evolutionary computation is designed and implemented in this paper to solve this problem. The experimental results demonstrate the feasibility of the optimization algorithm.展开更多
Optical microscopy is commonly used for cancer cell detection. Focusing on carcinoma cell identification via optical microscopy, a proof-of-concept study was performed at Laboratory of Design, Optimization and Modelin...Optical microscopy is commonly used for cancer cell detection. Focusing on carcinoma cell identification via optical microscopy, a proof-of-concept study was performed at Laboratory of Design, Optimization and Modeling (LCOMS) to determine the grade of cancer cells. This paper focuses on three types of abnormal cells;namely, Benign Hyperplasia (BH), Intraepithelial Neoplasia (IN), which is a precursor state for cancer, and Carcinoma (Ca), which corresponds to abnormal tissue proliferation cancer. These types of cells were used to assess the efficiency of using shape features to identify carcinoma cells. A comparative study based on performance indicator concludes that three features, Area, Xor-Convex, and Solidity, were found to be effective in identifying the Carcinoma grade of cancer cells.展开更多
A modified Fourier descriptor was presented. Information from a local space can be used more efficiently. After the boundary pixel set of an object was computed, centroid distance approach was used to compute shape si...A modified Fourier descriptor was presented. Information from a local space can be used more efficiently. After the boundary pixel set of an object was computed, centroid distance approach was used to compute shape signature in the local space. A pair of shape signature and boundary pixel gray was used as a point in a feature space. Then, Fourier transform was used for composition of point information in the feature space so that the shape features could be computed. It is proved theoretically that the shape features from modified Fourier descriptors are invariant to translation, rotation, scaling, and change of start point. It is also testified by measuring the retrieval performance of the systems that the shape features from modified Fourier oescriptors are more discriminative than those from other Fourier descriptors.展开更多
In the present study, it is expected to tailor the microstructural features, martensitic transformation temperatures and mechanical properties of Ti-V-Al shape memory alloys through adding Sn alloying elements, which ...In the present study, it is expected to tailor the microstructural features, martensitic transformation temperatures and mechanical properties of Ti-V-Al shape memory alloys through adding Sn alloying elements, which further expands their applications. Sn addition results in the monotonous rising of average valence electron number (e/a). In proportion, the single α″ martensite phase directly evolves into merely β parent phase in present Ti-V-Al-based shape memory alloys, as Sn content increases from 0.5 to 5.0 at.%. Meanwhile, Sn addition causes the reduction in the grain size. Combined with transmission electron microscopy (TEM) observation and d electron theory analysis, it can be speculated that Sn addition can suppress the precipitation of ω phase. With increasing Sn content, fracture strength invariably decreases from 962 to 792 MPa, whereas the yield strength firstly decreases and then increases. The lowest yield stress for the stress-induced martensitic transformation of 220 MPa can be obtained in Ti-V-Al shape memory alloy by adding 3.0 at.% Sn. By optimizing 1.0 at.% Sn, the excellent ductility with a largest elongation of 42.1% can be gained in Ti-V-Al shape memory alloy, which is larger than that of the reported Ti-V-Al-based shape memory alloys. Besides, as a result of solution strengthening and grain refinement, Ti-V-Al-based shape memory alloy with 5.0 at.% Sn possesses the highest yield strength, further contributing to the excellent strain recovery characteristics with 4% fully recoverable strain.展开更多
A new active shape models (ASMs) was presented, which is driven by scale invariant feature transform (SIFT) local descriptor instead of normalizing first order derivative profiles in the original formulation, to segme...A new active shape models (ASMs) was presented, which is driven by scale invariant feature transform (SIFT) local descriptor instead of normalizing first order derivative profiles in the original formulation, to segment lung fields from chest radiographs. The modified SIFT local descriptor, more distinctive than the general intensity and gradient features, is used to characterize the image features in the vicinity of each pixel at each resolution level during the segmentation optimization procedure. Experimental results show that the proposed method is more robust and accurate than the original ASMs in terms of an average overlap percentage and average contour distance in segmenting the lung fields from an available public database.展开更多
In recent years,the demand for synchronous acquisition of three-dimensional(3D)shape and col-or texture has surged in fields such as cultural heritage preservation and healthcare.Addressing this need,this paper propos...In recent years,the demand for synchronous acquisition of three-dimensional(3D)shape and col-or texture has surged in fields such as cultural heritage preservation and healthcare.Addressing this need,this paper proposes a novel method for simultaneous 3D shape and color texture capture.First,a linear model correlating camera exposure time with grayscale values is established.Through exposure time calibration,the projected red,green and blue(RGB)light and white-light grayscale values captured by a monochrome cam-era are aligned.Then,three sets of color fringes are projected onto the object to identify optimal pixels for 3D reconstruction.And,three pure-color patterns are projected to synthesize the color texture.Experimental res-ults show that this method effectively achieves synchronous 3D shape and color texture acquisition,offering high speed and precision,and avoids color crosstalk interference common in 3D reconstruction of colored ob-jects using a monochrome camera.展开更多
In order to accurately measure an object’s three-dimensional surface shape,the influence of sampling on it was studied.First,on the basis of deriving spectra expressions through the Fourier transform,the generation o...In order to accurately measure an object’s three-dimensional surface shape,the influence of sampling on it was studied.First,on the basis of deriving spectra expressions through the Fourier transform,the generation of CCD pixels was analyzed,and its expression was given.Then,based on the discrete expression of deformation fringes obtained after sampling,its Fourier spectrum expression was derived,resulting in an infinitely repeated"spectra island"in the frequency domain.Finally,on the basis of using a low-pass filter to remove high-order harmonic components and retaining only one fundamental frequency component,the inverse Fourier transform was used to reconstruct the signal strength.A method of reducing the sampling interval,i.e.,reducing the number of sampling points per fringe,was proposed to increase the ratio between the sampling frequency and the fundamental frequency of the grating.This was done to reconstruct the object’s surface shape more accurately under the condition of m>4.The basic principle was verified through simulation and experiment.In the simulation,the sampling intervals were 8 pixels,4 pixels,2 pixels,and 1 pixel,the maximum absolute error values obtained in the last three situations were 88.80%,38.38%,and 31.50%in the first situation,respectively,and the corresponding average absolute error values are 71.84%,43.27%,and 32.26%.It is demonstrated that the smaller the sampling interval,the better the recovery effect.Taking the same four sampling intervals in the experiment as in the simulation can also lead to the same conclusions.The simulated and experimental results show that reducing the sampling interval can improve the accuracy of object surface shape measurement and achieve better reconstruction results.展开更多
In optical three-dimensional shape measurement, a method of improving the measurement precision for phase reconstruction without phase unwrapping is analyzed in detail. Intensities of any five consecutive pixels that ...In optical three-dimensional shape measurement, a method of improving the measurement precision for phase reconstruction without phase unwrapping is analyzed in detail. Intensities of any five consecutive pixels that lie in the x-axis direction of the phase domain are given. Partial derivatives of the phase function in the x- and y-axis directions are obtained with a phase-shifting mechanism, the origin of which is analysed. Furthermore, to avoid phase unwrapping in the phase reconstruction, we derive the gradient of the phase function and perform a two-dimensional integral along the x- and y-axis directions. The reconstructed phase can be obtained directly by performing numerical integration, and thus it is of great convenience for phase reconstruction. Finally, the results of numerical simulations and practical experiments verify the correctness of the proposed method.展开更多
Three-dimensional(3D)shape registration is a challenging problem,especially for shapes under non-rigid transformations.In this paper,a 3D non-rigid shape registration method is proposed,called balanced functional maps...Three-dimensional(3D)shape registration is a challenging problem,especially for shapes under non-rigid transformations.In this paper,a 3D non-rigid shape registration method is proposed,called balanced functional maps(BFM).The BFM algorithm generalizes the point-based correspondence to functions.By choosing the Laplace-Beltrami eigenfunctions as the function basis,the transformations between shapes can be represented by the functional map(FM)matrix.In addition,many constraints on shape registration,such as the feature descriptor,keypoint,and salient region correspondence,can be formulated linearly using the matrix.By bi-directionally searching for the nearest neighbors of points’indicator functions in the function space,the point-based correspondence can be derived from FMs.We conducted several experiments on the Topology and Orchestration Specification for Cloud Applications(TOSCA)dataset and the Shape Completion and Animation of People(SCAPE)dataset.Experimental results show that the proposed BFM algorithm is effective and has superior performance than the state-of-the-art methods on both datasets.展开更多
In geometry processing,symmetry research benefits from global geo-metric features of complete shapes,but the shape of an object captured in real-world applications is often incomplete due to the limited sensor resoluti...In geometry processing,symmetry research benefits from global geo-metric features of complete shapes,but the shape of an object captured in real-world applications is often incomplete due to the limited sensor resolution,single viewpoint,and occlusion.Different from the existing works predicting symmetry from the complete shape,we propose a learning approach for symmetry predic-tion based on a single RGB-D image.Instead of directly predicting the symmetry from incomplete shapes,our method consists of two modules,i.e.,the multi-mod-al feature fusion module and the detection-by-reconstruction module.Firstly,we build a channel-transformer network(CTN)to extract cross-fusion features from the RGB-D as the multi-modal feature fusion module,which helps us aggregate features from the color and the depth separately.Then,our self-reconstruction net-work based on a 3D variational auto-encoder(3D-VAE)takes the global geo-metric features as input,followed by a prediction symmetry network to detect the symmetry.Our experiments are conducted on three public datasets:ShapeNet,YCB,and ScanNet,we demonstrate that our method can produce reliable and accurate results.展开更多
Traffic sign detection is a critical component of driving systems.Single-stage network-based traffic sign detection algorithms,renowned for their fast detection speeds and high accuracy,have become the dominant approa...Traffic sign detection is a critical component of driving systems.Single-stage network-based traffic sign detection algorithms,renowned for their fast detection speeds and high accuracy,have become the dominant approach in current practices.However,in complex and dynamic traffic scenes,particularly with smaller traffic sign objects,challenges such as missed and false detections can lead to reduced overall detection accuracy.To address this issue,this paper proposes a detection algorithm that integrates edge and shape information.Recognizing that traffic signs have specific shapes and distinct edge contours,this paper introduces an edge feature extraction branch within the backbone network,enabling adaptive fusion with features of the same hierarchical level.Additionally,a shape prior convolution module is designed to replaces the first two convolutional modules of the backbone network,aimed at enhancing the model's perception ability for specific shape objects and reducing its sensitivity to background noise.The algorithm was evaluated on the CCTSDB and TT100k datasets,and compared to YOLOv8s,the mAP50 values increased by 3.0%and 10.4%,respectively,demonstrating the effectiveness of the proposed method in improving the accuracy of traffic sign detection.展开更多
Active Shape Model （ASM） is a powerful statistical tool to extract the facial features of a face image under frontal view. It mainly relies on Principle Component Analysis （PCA） to statistically model the variabil...Active Shape Model （ASM） is a powerful statistical tool to extract the facial features of a face image under frontal view. It mainly relies on Principle Component Analysis （PCA） to statistically model the variability in the training set of example shapes. Independent Component Analysis （ICA） has been proven to be more efficient to extract face features than PCA. In this paper, we combine the PCA and ICA by the consecutive strategy to form a novel ASM. Firstly, an initial model, which shows the global shape variability in the training set, is generated by the PCA-based ASM. And then, the final shape model, which contains more local characters, is established by the ICA-based ASM. Experimental results verify that the accuracy of facial feature extraction is statistically significantly improved by applying the ICA modes after the PCA modes.展开更多
In expression recognition, feature representation is critical for successful recognition since it contains distinctive information of expressions. In this paper, a new approach for representing facial expression featu...In expression recognition, feature representation is critical for successful recognition since it contains distinctive information of expressions. In this paper, a new approach for representing facial expression features is proposed with its objective to describe features in an effective and efficient way in order to improve the recognition performance. The method combines the facial action coding system(FACS) and 'uniform' local binary patterns(LBP) to represent facial expression features from coarse to fine. The facial feature regions are extracted by active shape models(ASM) based on FACS to obtain the gray-level texture. Then, LBP is used to represent expression features for enhancing the discriminant. A facial expression recognition system is developed based on this feature extraction method by using K nearest neighborhood(K-NN) classifier to recognize facial expressions. Finally, experiments are carried out to evaluate this feature extraction method. The significance of removing the unrelated facial regions and enhancing the discrimination ability of expression features in the recognition process is indicated by the results, in addition to its convenience.展开更多
In feature based image matching,distinctive features in images are detected and represented by feature descriptors.Matching is then carried out by assessing the similarity of the descriptors of potentially conjugate p...In feature based image matching,distinctive features in images are detected and represented by feature descriptors.Matching is then carried out by assessing the similarity of the descriptors of potentially conjugate points.In this paper,we first shortly discuss the general frame-work.Then,we review feature detection as well as the determination of affine shape and orientation of local features,before analyzing feature description in more detail.In the feature description review,the general framework of local feature description is presented first.Then,the review discusses the evolution from hand-crafted feature descriptors,e.g.SIFT(Scale Invariant Feature Transform),to machine learning and deep learning based descriptors.The machine learning models,the training loss and the respective training data of learning-based algorithms are looked at in more detail;subsequently the various advantages and challenges of the different approaches are discussed.Finally,we present and assess some current research directions before concluding the paper.展开更多
基金Supported by the National Key R&D Program of China(2022YFC3803600).
文摘Background Three-dimensional(3D)shape representation using mesh data is essential in various applications,such as virtual reality and simulation technologies.Current methods for extracting features from mesh edges or faces struggle with complex 3D models because edge-based approaches miss global contexts and face-based methods overlook variations in adjacent areas,which affects the overall precision.To address these issues,we propose the Feature Discrimination and Context Propagation Network(FDCPNet),which is a novel approach that synergistically integrates local and global features in mesh datasets.Methods FDCPNet is composed of two modules:(1)the Feature Discrimination Module,which employs an attention mechanism to enhance the identification of key local features,and(2)the Context Propagation Module,which enriches key local features by integrating global contextual information,thereby facilitating a more detailed and comprehensive representation of crucial areas within the mesh model.Results Experiments on popular datasets validated the effectiveness of FDCPNet,showing an improvement in the classification accuracy over the baseline MeshNet.Furthermore,even with reduced mesh face numbers and limited training data,FDCPNet achieved promising results,demonstrating its robustness in scenarios of variable complexity.
基金funded by the Natural Science Foundation of China (Grant nos. 41674085 and 41621091)the National Key Basic Research Program of China (973 program, Grant nos. 2012CB957703 and 2013CB733301)
文摘To estimate basal water storage beneath the Antarctic ice sheet, it is essential to have data on the three-dimensional characteristics of subglacial lakes. We present a method to estimate the water depth and surface area of Antarctic subglacial lakes from the inversion of hydraulic potential method. Lake Vostok is chosen as a case study because of the diverse and comprehensive measurements that have been obtained over and around the lake. The average depth of Lake Vostok is around 345±4 m. We estimated the surface area of Lake Vostok beneath the ice sheet to be about 13300±594 km^2. The lake consists of two sub-basins separated by a ridge at water depths of about 200–300 m. The surface area of the northern sub-basin is estimated to be about half of that of the southern basin. The maximum depths of the northern and southern sub-basins are estimated to be about 450 and 850 m, respectively. Total water volume is estimated to be about 4658±204 km^3. These estimates are compared with previous estimates obtained from seismic data and inversion of aerogravity data. In general, our estimates are closer to those obtained from the inversion of aerogravity data than those from seismic data, indicating the applicability of our method to the estimation of water depths of other subglacial lakes.
基金Project supported by the Zhejiang Provincial Welfare Technology Applied Research Project,China(Grant No.2017C31080)
文摘Dithering optimization techniques can be divided into the phase-optimized technique and the intensity-optimized technique. The problem with the former is the poor sensitivity to various defocusing amounts, and the problem with the latter is that it cannot enhance phase quality directly nor efficiently. In this paper, we present a multi-objective optimization framework for three-dimensional(3D) measurement by utilizing binary defocusing technique. Moreover, a binary patch optimization technique is used to solve the time-consuming issue of genetic algorithm. It is demonstrated that the presented technique consistently obtains significant phase performance improvement under various defocusing amounts.
基金the National Natural Science Foundation of China (60303029)
文摘The image shape feature can be described by the image Zernike moments. In this paper, we points out the problem that the high dimension image Zernike moments shape feature vector can describe more detail of the original image but has too many elements making trouble for the next image analysis phases. Then the low dimension image Zernike moments shape feature vector should be improved and optimized to describe more detail of the original image. So the optimization algorithm based on evolutionary computation is designed and implemented in this paper to solve this problem. The experimental results demonstrate the feasibility of the optimization algorithm.
文摘Optical microscopy is commonly used for cancer cell detection. Focusing on carcinoma cell identification via optical microscopy, a proof-of-concept study was performed at Laboratory of Design, Optimization and Modeling (LCOMS) to determine the grade of cancer cells. This paper focuses on three types of abnormal cells;namely, Benign Hyperplasia (BH), Intraepithelial Neoplasia (IN), which is a precursor state for cancer, and Carcinoma (Ca), which corresponds to abnormal tissue proliferation cancer. These types of cells were used to assess the efficiency of using shape features to identify carcinoma cells. A comparative study based on performance indicator concludes that three features, Area, Xor-Convex, and Solidity, were found to be effective in identifying the Carcinoma grade of cancer cells.
基金Project(60873010)supported by the National Natural Science Foundation of ChinaProject supported by the Doctor Startup Foundation of Shenyang University of Technology,China
文摘A modified Fourier descriptor was presented. Information from a local space can be used more efficiently. After the boundary pixel set of an object was computed, centroid distance approach was used to compute shape signature in the local space. A pair of shape signature and boundary pixel gray was used as a point in a feature space. Then, Fourier transform was used for composition of point information in the feature space so that the shape features could be computed. It is proved theoretically that the shape features from modified Fourier descriptors are invariant to translation, rotation, scaling, and change of start point. It is also testified by measuring the retrieval performance of the systems that the shape features from modified Fourier oescriptors are more discriminative than those from other Fourier descriptors.
基金financial support from the National Natural Science Foundation of China(Nos.52101231,52101232 and 51871079)the Science Fund of Shandong Laboratory of Advanced Materials and Green Manufacturing(Yantai)(No.AMGM2021F09)+1 种基金the Natural Science Foundation of Shandong Province,China(No.ZR2021QE044)the Gansu Province Science and Technology Foundation for Youths(No.21JR7RA088).
文摘In the present study, it is expected to tailor the microstructural features, martensitic transformation temperatures and mechanical properties of Ti-V-Al shape memory alloys through adding Sn alloying elements, which further expands their applications. Sn addition results in the monotonous rising of average valence electron number (e/a). In proportion, the single α″ martensite phase directly evolves into merely β parent phase in present Ti-V-Al-based shape memory alloys, as Sn content increases from 0.5 to 5.0 at.%. Meanwhile, Sn addition causes the reduction in the grain size. Combined with transmission electron microscopy (TEM) observation and d electron theory analysis, it can be speculated that Sn addition can suppress the precipitation of ω phase. With increasing Sn content, fracture strength invariably decreases from 962 to 792 MPa, whereas the yield strength firstly decreases and then increases. The lowest yield stress for the stress-induced martensitic transformation of 220 MPa can be obtained in Ti-V-Al shape memory alloy by adding 3.0 at.% Sn. By optimizing 1.0 at.% Sn, the excellent ductility with a largest elongation of 42.1% can be gained in Ti-V-Al shape memory alloy, which is larger than that of the reported Ti-V-Al-based shape memory alloys. Besides, as a result of solution strengthening and grain refinement, Ti-V-Al-based shape memory alloy with 5.0 at.% Sn possesses the highest yield strength, further contributing to the excellent strain recovery characteristics with 4% fully recoverable strain.
基金The National Natural Science Foundation of China(No60271033)
文摘A new active shape models (ASMs) was presented, which is driven by scale invariant feature transform (SIFT) local descriptor instead of normalizing first order derivative profiles in the original formulation, to segment lung fields from chest radiographs. The modified SIFT local descriptor, more distinctive than the general intensity and gradient features, is used to characterize the image features in the vicinity of each pixel at each resolution level during the segmentation optimization procedure. Experimental results show that the proposed method is more robust and accurate than the original ASMs in terms of an average overlap percentage and average contour distance in segmenting the lung fields from an available public database.
文摘In recent years,the demand for synchronous acquisition of three-dimensional(3D)shape and col-or texture has surged in fields such as cultural heritage preservation and healthcare.Addressing this need,this paper proposes a novel method for simultaneous 3D shape and color texture capture.First,a linear model correlating camera exposure time with grayscale values is established.Through exposure time calibration,the projected red,green and blue(RGB)light and white-light grayscale values captured by a monochrome cam-era are aligned.Then,three sets of color fringes are projected onto the object to identify optimal pixels for 3D reconstruction.And,three pure-color patterns are projected to synthesize the color texture.Experimental res-ults show that this method effectively achieves synchronous 3D shape and color texture acquisition,offering high speed and precision,and avoids color crosstalk interference common in 3D reconstruction of colored ob-jects using a monochrome camera.
文摘In order to accurately measure an object’s three-dimensional surface shape,the influence of sampling on it was studied.First,on the basis of deriving spectra expressions through the Fourier transform,the generation of CCD pixels was analyzed,and its expression was given.Then,based on the discrete expression of deformation fringes obtained after sampling,its Fourier spectrum expression was derived,resulting in an infinitely repeated"spectra island"in the frequency domain.Finally,on the basis of using a low-pass filter to remove high-order harmonic components and retaining only one fundamental frequency component,the inverse Fourier transform was used to reconstruct the signal strength.A method of reducing the sampling interval,i.e.,reducing the number of sampling points per fringe,was proposed to increase the ratio between the sampling frequency and the fundamental frequency of the grating.This was done to reconstruct the object’s surface shape more accurately under the condition of m>4.The basic principle was verified through simulation and experiment.In the simulation,the sampling intervals were 8 pixels,4 pixels,2 pixels,and 1 pixel,the maximum absolute error values obtained in the last three situations were 88.80%,38.38%,and 31.50%in the first situation,respectively,and the corresponding average absolute error values are 71.84%,43.27%,and 32.26%.It is demonstrated that the smaller the sampling interval,the better the recovery effect.Taking the same four sampling intervals in the experiment as in the simulation can also lead to the same conclusions.The simulated and experimental results show that reducing the sampling interval can improve the accuracy of object surface shape measurement and achieve better reconstruction results.
基金Project supported by the National Natural Science Foundation of China (Grant No. 61144006)
文摘In optical three-dimensional shape measurement, a method of improving the measurement precision for phase reconstruction without phase unwrapping is analyzed in detail. Intensities of any five consecutive pixels that lie in the x-axis direction of the phase domain are given. Partial derivatives of the phase function in the x- and y-axis directions are obtained with a phase-shifting mechanism, the origin of which is analysed. Furthermore, to avoid phase unwrapping in the phase reconstruction, we derive the gradient of the phase function and perform a two-dimensional integral along the x- and y-axis directions. The reconstructed phase can be obtained directly by performing numerical integration, and thus it is of great convenience for phase reconstruction. Finally, the results of numerical simulations and practical experiments verify the correctness of the proposed method.
基金the China Scholarship Council under Grant No.201406070059.
文摘Three-dimensional(3D)shape registration is a challenging problem,especially for shapes under non-rigid transformations.In this paper,a 3D non-rigid shape registration method is proposed,called balanced functional maps(BFM).The BFM algorithm generalizes the point-based correspondence to functions.By choosing the Laplace-Beltrami eigenfunctions as the function basis,the transformations between shapes can be represented by the functional map(FM)matrix.In addition,many constraints on shape registration,such as the feature descriptor,keypoint,and salient region correspondence,can be formulated linearly using the matrix.By bi-directionally searching for the nearest neighbors of points’indicator functions in the function space,the point-based correspondence can be derived from FMs.We conducted several experiments on the Topology and Orchestration Specification for Cloud Applications(TOSCA)dataset and the Shape Completion and Animation of People(SCAPE)dataset.Experimental results show that the proposed BFM algorithm is effective and has superior performance than the state-of-the-art methods on both datasets.
文摘In geometry processing,symmetry research benefits from global geo-metric features of complete shapes,but the shape of an object captured in real-world applications is often incomplete due to the limited sensor resolution,single viewpoint,and occlusion.Different from the existing works predicting symmetry from the complete shape,we propose a learning approach for symmetry predic-tion based on a single RGB-D image.Instead of directly predicting the symmetry from incomplete shapes,our method consists of two modules,i.e.,the multi-mod-al feature fusion module and the detection-by-reconstruction module.Firstly,we build a channel-transformer network(CTN)to extract cross-fusion features from the RGB-D as the multi-modal feature fusion module,which helps us aggregate features from the color and the depth separately.Then,our self-reconstruction net-work based on a 3D variational auto-encoder(3D-VAE)takes the global geo-metric features as input,followed by a prediction symmetry network to detect the symmetry.Our experiments are conducted on three public datasets:ShapeNet,YCB,and ScanNet,we demonstrate that our method can produce reliable and accurate results.
基金supported by the National Natural Science Foundation of China(Grant Nos.62572057,62272049,U24A20331)Beijing Natural Science Foundation(Grant Nos.4232026,4242020)Academic Research Projects of Beijing Union University(Grant No.ZK10202404).
文摘Traffic sign detection is a critical component of driving systems.Single-stage network-based traffic sign detection algorithms,renowned for their fast detection speeds and high accuracy,have become the dominant approach in current practices.However,in complex and dynamic traffic scenes,particularly with smaller traffic sign objects,challenges such as missed and false detections can lead to reduced overall detection accuracy.To address this issue,this paper proposes a detection algorithm that integrates edge and shape information.Recognizing that traffic signs have specific shapes and distinct edge contours,this paper introduces an edge feature extraction branch within the backbone network,enabling adaptive fusion with features of the same hierarchical level.Additionally,a shape prior convolution module is designed to replaces the first two convolutional modules of the backbone network,aimed at enhancing the model's perception ability for specific shape objects and reducing its sensitivity to background noise.The algorithm was evaluated on the CCTSDB and TT100k datasets,and compared to YOLOv8s,the mAP50 values increased by 3.0%and 10.4%,respectively,demonstrating the effectiveness of the proposed method in improving the accuracy of traffic sign detection.
文摘Active Shape Model （ASM） is a powerful statistical tool to extract the facial features of a face image under frontal view. It mainly relies on Principle Component Analysis （PCA） to statistically model the variability in the training set of example shapes. Independent Component Analysis （ICA） has been proven to be more efficient to extract face features than PCA. In this paper, we combine the PCA and ICA by the consecutive strategy to form a novel ASM. Firstly, an initial model, which shows the global shape variability in the training set, is generated by the PCA-based ASM. And then, the final shape model, which contains more local characters, is established by the ICA-based ASM. Experimental results verify that the accuracy of facial feature extraction is statistically significantly improved by applying the ICA modes after the PCA modes.
基金supported by National Natural Science Foundation of China(No.61273339)
文摘In expression recognition, feature representation is critical for successful recognition since it contains distinctive information of expressions. In this paper, a new approach for representing facial expression features is proposed with its objective to describe features in an effective and efficient way in order to improve the recognition performance. The method combines the facial action coding system(FACS) and 'uniform' local binary patterns(LBP) to represent facial expression features from coarse to fine. The facial feature regions are extracted by active shape models(ASM) based on FACS to obtain the gray-level texture. Then, LBP is used to represent expression features for enhancing the discriminant. A facial expression recognition system is developed based on this feature extraction method by using K nearest neighborhood(K-NN) classifier to recognize facial expressions. Finally, experiments are carried out to evaluate this feature extraction method. The significance of removing the unrelated facial regions and enhancing the discrimination ability of expression features in the recognition process is indicated by the results, in addition to its convenience.
基金The authors would like to thank NVIDIA Corp.for donating the GPU used in this research through its GPU grant program.The first author Lin Chen would also like to thank the China Scholarship Council(CSC)for financially supporting his PhD study.
文摘In feature based image matching,distinctive features in images are detected and represented by feature descriptors.Matching is then carried out by assessing the similarity of the descriptors of potentially conjugate points.In this paper,we first shortly discuss the general frame-work.Then,we review feature detection as well as the determination of affine shape and orientation of local features,before analyzing feature description in more detail.In the feature description review,the general framework of local feature description is presented first.Then,the review discusses the evolution from hand-crafted feature descriptors,e.g.SIFT(Scale Invariant Feature Transform),to machine learning and deep learning based descriptors.The machine learning models,the training loss and the respective training data of learning-based algorithms are looked at in more detail;subsequently the various advantages and challenges of the different approaches are discussed.Finally,we present and assess some current research directions before concluding the paper.
