目的近年来,由于局部图像描述符在大的视角与光度变化、噪声、局部遮挡等方面具有良好性能,已成功应用于图像搜索、机器人导航、图像分类、视频行为识别等各种计算机视觉研究领域。方法提出了一种新的用于图像区域描述的局部特征:局部...目的近年来,由于局部图像描述符在大的视角与光度变化、噪声、局部遮挡等方面具有良好性能,已成功应用于图像搜索、机器人导航、图像分类、视频行为识别等各种计算机视觉研究领域。方法提出了一种新的用于图像区域描述的局部特征:局部灰度极值模式(LIEP)。在离一个像素点半径不同的两个同心圆上分别均匀抽样相同点数的采样点,不同同心圆上采样点与中心像素点之间的夹角相互内插,分别独立计算每个同心圆上采样点的最大和最小灰度模式。计算半径小的同心圆上的最大灰度模式和半径大的同心圆上的最小灰度模式的2维联合分布,得到一种极值模式。再计算半径小的同心圆上的最小灰度模式和半径大的同心圆上的最大灰度模式的2维联合分布,得到另一种极值模式。最后对这2种极值模式进行级联,得到LIEP。相对于局部灰度序模式和局部二进制模式,LIEP在图像光度和几何变化下更稳定,抗噪声性能更强,出现模式错误的概率更小。LIEP在局部旋转不变坐标系统下计算,采用多支撑域和图像块全局灰度序空间汇聚方法得到一种新的局部图像描述符:LIEP空间分布直方图(LIEPH)。LIEPH描述符具有单调光照不变性和在不计算图像块主方向条件下保持旋转不变性。结果在标准图像匹配数据库上的实验表明:LIEPH的查全率-查错率曲线都位于最上方,匹配性能大大优于单支撑域描述符SIFT(scale invariant feature transform)、CS-LBP(center-symmetric local binary pattern)、LIOP(local intensity order pattern)、HRI-CSLTP(histogram of relative intensities and center-symmetric local ternary patterns)、EOD(exact order based descriptor)及多支撑域描述符MRRID(multisupport region rotation and intensity monotonic imariant descriptor)。在大的图像几何畸变下,LIEPH更能展现优越的匹配性能。在对描述符进行定量分析的实验中,当查错率(1-preci展开更多
This paper presents an efficient image feature representation method, namely angle structure descriptor(ASD), which is built based on the angle structures of images. According to the diversity in directions, angle str...This paper presents an efficient image feature representation method, namely angle structure descriptor(ASD), which is built based on the angle structures of images. According to the diversity in directions, angle structures are defined in local blocks. Combining color information in HSV color space, we use angle structures to detect images. The internal correlations between neighboring pixels in angle structures are explored to form a feature vector. With angle structures as bridges, ASD extracts image features by integrating multiple information as a whole, such as color, texture, shape and spatial layout information. In addition, the proposed algorithm is efficient for image retrieval without any clustering implementation or model training. Experimental results demonstrate that ASD outperforms the other related algorithms.展开更多
This paper presents a biologically inspired local image descriptor that combines color and shape features. Compared with previous descriptors, red-cyan cells associated with L, M, and S cones （L for long, M for mediu...This paper presents a biologically inspired local image descriptor that combines color and shape features. Compared with previous descriptors, red-cyan cells associated with L, M, and S cones （L for long, M for medium, and S for short） are used to indicate one of the opponent color channels. Stepping forward from state-of-the-art color feature extraction, we exploit a new approach to compute the color orientation and magnitudes of three opponent color channels, namely, red-green, blue-yellow, and red-cyan, in two-dimensional space. Color orientation is calculated in histograms with magnitude weighting. We linearly concatenate the four-color-opponent-channel histogram and scale-invariant-feamre-transform histogram in the final step. We apply our biologically inspired descriptor to describe the local image feature. Quantitative comparisons with state-of-the-art descriptors demonstrate the significant advantages of maintaining invariance to photometric and geometric changes in image matching, particularly in cases, such as illumination variation and image blurring, where more color contrast information is observed.展开更多
The extraction and description of image features are very important for visual simultaneous localization and mapping(V-SLAM).A rotated boosted efficient binary local image descriptor(BEBLID)SLAM(RB-SLAM)algorithm base...The extraction and description of image features are very important for visual simultaneous localization and mapping(V-SLAM).A rotated boosted efficient binary local image descriptor(BEBLID)SLAM(RB-SLAM)algorithm based on improved oriented fast and rotated brief(ORB)feature description is proposed in this paper,which can solve the problems of low localization accuracy and time efficiency of the current ORB-SLAM3 algorithm.Firstly,it uses the BEBLID to replace the feature point description algorithm of the original ORB to enhance the expressiveness and description efficiency of the image.Secondly,it adds rotational invariance to the BEBLID using the orientation information of the feature points.It also selects the rotationally stable bits in the BEBLID to further enhance the rotational invariance of the BEBLID.Finally,it retrains the binary visual dictionary based on the BEBLID to reduce the cumulative error of V-SLAM and improve the loading speed of the visual dictionary.Experiments show that the dictionary loading efficiency is improved by more than 10 times.The RB-SLAM algorithm improves the trajectory accuracy by 24.75%on the TUM dataset and 26.25%on the EuRoC dataset compared to the ORB-SLAM3 algorithm.展开更多
提出了一种新的局部图像描述符:特征联合和旋转不变空间分割联合描述符(Feature combination and rotation invariant space division combination descriptor,FCSCD).提出了一种新的局部特征:WLBP(Weber local binary pattern),该特征...提出了一种新的局部图像描述符:特征联合和旋转不变空间分割联合描述符(Feature combination and rotation invariant space division combination descriptor,FCSCD).提出了一种新的局部特征:WLBP(Weber local binary pattern),该特征由局部二进制模式和韦伯二进制差分激励联合得到.提出了一种新的用于特征汇聚的旋转不变空间分割方法,该方法由强度序空间分割和圆环空间分割联合得到.WLBP在局部旋转不变坐标系计算得到,强度序和圆环空间分割本身也具有旋转不变性,所以FCSCD描述符在不需要计算图像块主方向下保持了旋转不变性.与现有的局部描述符相比,本文的联合方法编码了多种类型的信息在描述符直方图中,所以FCSCD辨别能力更强,鲁棒性更强.图像匹配实验结果表明了本文方法的有效性和优越性,所提出的描述符具有很高的匹配性能,优于其他的主流局部描述符(SIFT、CS-LBP、OSID、LIOP、EOD和MRRID).展开更多
A discriminative local shape descriptor plays an important role in various applications.In this paper,we present a novel deep learning framework that derives discriminative local descriptors for deformable 3D shapes.W...A discriminative local shape descriptor plays an important role in various applications.In this paper,we present a novel deep learning framework that derives discriminative local descriptors for deformable 3D shapes.We use local"geometry images"to encode the multi-scale local features of a point,via an intrinsic parameterization method based on geodesic polar coordinates.This new parameterization provides robust geometry images even for badly-shaped triangular meshes.Then a triplet network with shared architecture and parameters is used to perform deep metric learning;its aim is to distinguish between similar and dissimilar pairs of points.Additionally,a newly designed triplet loss function is minimized for improved,accurate training of the triplet network.To solve the dense correspondence problem,an efficient sampling approach is utilized to achieve a good compromise between training performance and descriptor quality.During testing,given a geometry image of a point of interest,our network outputs a discriminative local descriptor for it.Extensive testing of non-rigid dense shape matching on a variety of benchmarks demonstrates the superiority of the proposed descriptors over the state-of-the-art alternatives.展开更多
目的局部图像描述符凭借其优越的特性广泛应用于计算机视觉和图像处理多个领域,如图像匹配、图像分类、图像搜索、从运动恢复结构等。方法本文提出了一种新的局部特征:最大边缘方向模式(MEOP)。该特征计算中心像素和周围像素间最大强度...目的局部图像描述符凭借其优越的特性广泛应用于计算机视觉和图像处理多个领域,如图像匹配、图像分类、图像搜索、从运动恢复结构等。方法本文提出了一种新的局部特征:最大边缘方向模式(MEOP)。该特征计算中心像素和周围像素间最大强度差值,对其位置和符号进行编码。呈现最大强度差值的像素代表局部领域的最强边缘处,其位置描述了径向方向,差值的符号描述了径向方向的朝向。相对于局部二进制模式,由于MEOP仅编码最大强度差值,所以只要最大强度差值的位置和符号不出现改变,MEOP模式就不会发生改变。所以MEOP模式的鲁棒性较高,抗噪声能力更强。MEOP在描述图像的局部结构特征上和局部二进制模式是完全不一样的,两种模式在表达图像的局部结构方面具有较大的互补性。利用局部旋转不变坐标系计算最大边缘方向模式,采用旋转不变强度序空间分割方法和多支撑域对最大边缘方向模式进行空间汇聚得到一种新的局部图像描述符:最大边缘方向模式直方图(MEOPH)。相对于采用局部二进制模式的MRRID(multisupport region rotation and intensity monotonic invariant descriptor)描述符相比,采用最大边缘方向模式的MEOPH描述符具有不同的统计特性和更优越的性能。结果在牛津大学仿射不变研究小组的标准测试图像集上对目前的主流局部描述符(SIFT(scale invariant feature transform)、DAISY、CS-LBP(center-symmetric local binary pattern)、HRI-CSLTP(histogram of relative intensities and center-symmetric local ternary patterns)和MRRID)进行了图像匹配实验。标准测试图像集上的实验结果表明,本文MEOPH和MRRID获得了最好的性能,MEOPH在所有测试数据集上的匹配性能都优于SIFT、DAISY、CS-LBP和HRI-CSLTP,在大多数情况下MEOPH的图像匹配效果要比MRRID稍好一些。在标准测试图像集上添加高斯噪声的图像匹配实验中,MEOPH�展开更多
针对图像匹配中SURF(speed up robust features)算法匹配效率不佳,以及RANSAC算法迭代次数不稳定和人为设置内点离散阈值所带来误差的问题,提出了一种结合改进的边缘化采样一致性算法和改进SURF的图像匹配方法。首先对输入图像进行快速...针对图像匹配中SURF(speed up robust features)算法匹配效率不佳,以及RANSAC算法迭代次数不稳定和人为设置内点离散阈值所带来误差的问题,提出了一种结合改进的边缘化采样一致性算法和改进SURF的图像匹配方法。首先对输入图像进行快速引导滤波预处理,过滤图像噪声并保留边缘细节信息。然后通过BEBLID(Boosted Efficient Binary Local Image Descriptor)算法为SURF构建高效的二值描述符,结合改进的边缘化采样一致性算法边缘化外点去除误匹配。经实验对比,该方法相较于SURF准确性更高,实时性有较大提升,可满足多数复杂环境下的图像匹配。展开更多
文摘目的近年来,由于局部图像描述符在大的视角与光度变化、噪声、局部遮挡等方面具有良好性能,已成功应用于图像搜索、机器人导航、图像分类、视频行为识别等各种计算机视觉研究领域。方法提出了一种新的用于图像区域描述的局部特征:局部灰度极值模式(LIEP)。在离一个像素点半径不同的两个同心圆上分别均匀抽样相同点数的采样点,不同同心圆上采样点与中心像素点之间的夹角相互内插,分别独立计算每个同心圆上采样点的最大和最小灰度模式。计算半径小的同心圆上的最大灰度模式和半径大的同心圆上的最小灰度模式的2维联合分布,得到一种极值模式。再计算半径小的同心圆上的最小灰度模式和半径大的同心圆上的最大灰度模式的2维联合分布,得到另一种极值模式。最后对这2种极值模式进行级联,得到LIEP。相对于局部灰度序模式和局部二进制模式,LIEP在图像光度和几何变化下更稳定,抗噪声性能更强,出现模式错误的概率更小。LIEP在局部旋转不变坐标系统下计算,采用多支撑域和图像块全局灰度序空间汇聚方法得到一种新的局部图像描述符:LIEP空间分布直方图(LIEPH)。LIEPH描述符具有单调光照不变性和在不计算图像块主方向条件下保持旋转不变性。结果在标准图像匹配数据库上的实验表明:LIEPH的查全率-查错率曲线都位于最上方,匹配性能大大优于单支撑域描述符SIFT(scale invariant feature transform)、CS-LBP(center-symmetric local binary pattern)、LIOP(local intensity order pattern)、HRI-CSLTP(histogram of relative intensities and center-symmetric local ternary patterns)、EOD(exact order based descriptor)及多支撑域描述符MRRID(multisupport region rotation and intensity monotonic imariant descriptor)。在大的图像几何畸变下,LIEPH更能展现优越的匹配性能。在对描述符进行定量分析的实验中,当查错率(1-preci
基金supported by the National Natural Science Foundation of China (No.61170145, 61373081, 61402268, 61401260, 61572298)the Technology and Development Project of Shandong (No.2013GGX10125)+1 种基金the Natural Science Foundation of Shandong China (No.BS2014DX006, ZR2014FM012)the Taishan Scholar Project of Shandong, China
文摘This paper presents an efficient image feature representation method, namely angle structure descriptor(ASD), which is built based on the angle structures of images. According to the diversity in directions, angle structures are defined in local blocks. Combining color information in HSV color space, we use angle structures to detect images. The internal correlations between neighboring pixels in angle structures are explored to form a feature vector. With angle structures as bridges, ASD extracts image features by integrating multiple information as a whole, such as color, texture, shape and spatial layout information. In addition, the proposed algorithm is efficient for image retrieval without any clustering implementation or model training. Experimental results demonstrate that ASD outperforms the other related algorithms.
基金Acknowledgment This study was supported by the National Natural Science Foundation of China （grant 61101155） and the Jilin Province Science and Technology Development Program （20101504）.
文摘This paper presents a biologically inspired local image descriptor that combines color and shape features. Compared with previous descriptors, red-cyan cells associated with L, M, and S cones （L for long, M for medium, and S for short） are used to indicate one of the opponent color channels. Stepping forward from state-of-the-art color feature extraction, we exploit a new approach to compute the color orientation and magnitudes of three opponent color channels, namely, red-green, blue-yellow, and red-cyan, in two-dimensional space. Color orientation is calculated in histograms with magnitude weighting. We linearly concatenate the four-color-opponent-channel histogram and scale-invariant-feamre-transform histogram in the final step. We apply our biologically inspired descriptor to describe the local image feature. Quantitative comparisons with state-of-the-art descriptors demonstrate the significant advantages of maintaining invariance to photometric and geometric changes in image matching, particularly in cases, such as illumination variation and image blurring, where more color contrast information is observed.
文摘The extraction and description of image features are very important for visual simultaneous localization and mapping(V-SLAM).A rotated boosted efficient binary local image descriptor(BEBLID)SLAM(RB-SLAM)algorithm based on improved oriented fast and rotated brief(ORB)feature description is proposed in this paper,which can solve the problems of low localization accuracy and time efficiency of the current ORB-SLAM3 algorithm.Firstly,it uses the BEBLID to replace the feature point description algorithm of the original ORB to enhance the expressiveness and description efficiency of the image.Secondly,it adds rotational invariance to the BEBLID using the orientation information of the feature points.It also selects the rotationally stable bits in the BEBLID to further enhance the rotational invariance of the BEBLID.Finally,it retrains the binary visual dictionary based on the BEBLID to reduce the cumulative error of V-SLAM and improve the loading speed of the visual dictionary.Experiments show that the dictionary loading efficiency is improved by more than 10 times.The RB-SLAM algorithm improves the trajectory accuracy by 24.75%on the TUM dataset and 26.25%on the EuRoC dataset compared to the ORB-SLAM3 algorithm.
文摘提出了一种新的局部图像描述符:特征联合和旋转不变空间分割联合描述符(Feature combination and rotation invariant space division combination descriptor,FCSCD).提出了一种新的局部特征:WLBP(Weber local binary pattern),该特征由局部二进制模式和韦伯二进制差分激励联合得到.提出了一种新的用于特征汇聚的旋转不变空间分割方法,该方法由强度序空间分割和圆环空间分割联合得到.WLBP在局部旋转不变坐标系计算得到,强度序和圆环空间分割本身也具有旋转不变性,所以FCSCD描述符在不需要计算图像块主方向下保持了旋转不变性.与现有的局部描述符相比,本文的联合方法编码了多种类型的信息在描述符直方图中,所以FCSCD辨别能力更强,鲁棒性更强.图像匹配实验结果表明了本文方法的有效性和优越性,所提出的描述符具有很高的匹配性能,优于其他的主流局部描述符(SIFT、CS-LBP、OSID、LIOP、EOD和MRRID).
基金partially funded by the National Key R&D Program of China(2018YFB2100602)the National Natural Science Foundation of China(61802406,61772523,61702488)+2 种基金Beijing Natural Science Foundation(L182059)the CCF–Tencent Open Research Fund,Shenzhen Basic Research Program(JCYJ20180507182222355)the Open Project Program of the State Key Lab of CAD&CG(A2004)Zhejiang University.
文摘A discriminative local shape descriptor plays an important role in various applications.In this paper,we present a novel deep learning framework that derives discriminative local descriptors for deformable 3D shapes.We use local"geometry images"to encode the multi-scale local features of a point,via an intrinsic parameterization method based on geodesic polar coordinates.This new parameterization provides robust geometry images even for badly-shaped triangular meshes.Then a triplet network with shared architecture and parameters is used to perform deep metric learning;its aim is to distinguish between similar and dissimilar pairs of points.Additionally,a newly designed triplet loss function is minimized for improved,accurate training of the triplet network.To solve the dense correspondence problem,an efficient sampling approach is utilized to achieve a good compromise between training performance and descriptor quality.During testing,given a geometry image of a point of interest,our network outputs a discriminative local descriptor for it.Extensive testing of non-rigid dense shape matching on a variety of benchmarks demonstrates the superiority of the proposed descriptors over the state-of-the-art alternatives.
文摘目的局部图像描述符凭借其优越的特性广泛应用于计算机视觉和图像处理多个领域,如图像匹配、图像分类、图像搜索、从运动恢复结构等。方法本文提出了一种新的局部特征:最大边缘方向模式(MEOP)。该特征计算中心像素和周围像素间最大强度差值,对其位置和符号进行编码。呈现最大强度差值的像素代表局部领域的最强边缘处,其位置描述了径向方向,差值的符号描述了径向方向的朝向。相对于局部二进制模式,由于MEOP仅编码最大强度差值,所以只要最大强度差值的位置和符号不出现改变,MEOP模式就不会发生改变。所以MEOP模式的鲁棒性较高,抗噪声能力更强。MEOP在描述图像的局部结构特征上和局部二进制模式是完全不一样的,两种模式在表达图像的局部结构方面具有较大的互补性。利用局部旋转不变坐标系计算最大边缘方向模式,采用旋转不变强度序空间分割方法和多支撑域对最大边缘方向模式进行空间汇聚得到一种新的局部图像描述符:最大边缘方向模式直方图(MEOPH)。相对于采用局部二进制模式的MRRID(multisupport region rotation and intensity monotonic invariant descriptor)描述符相比,采用最大边缘方向模式的MEOPH描述符具有不同的统计特性和更优越的性能。结果在牛津大学仿射不变研究小组的标准测试图像集上对目前的主流局部描述符(SIFT(scale invariant feature transform)、DAISY、CS-LBP(center-symmetric local binary pattern)、HRI-CSLTP(histogram of relative intensities and center-symmetric local ternary patterns)和MRRID)进行了图像匹配实验。标准测试图像集上的实验结果表明,本文MEOPH和MRRID获得了最好的性能,MEOPH在所有测试数据集上的匹配性能都优于SIFT、DAISY、CS-LBP和HRI-CSLTP,在大多数情况下MEOPH的图像匹配效果要比MRRID稍好一些。在标准测试图像集上添加高斯噪声的图像匹配实验中,MEOPH�
文摘针对图像匹配中SURF(speed up robust features)算法匹配效率不佳,以及RANSAC算法迭代次数不稳定和人为设置内点离散阈值所带来误差的问题,提出了一种结合改进的边缘化采样一致性算法和改进SURF的图像匹配方法。首先对输入图像进行快速引导滤波预处理,过滤图像噪声并保留边缘细节信息。然后通过BEBLID(Boosted Efficient Binary Local Image Descriptor)算法为SURF构建高效的二值描述符,结合改进的边缘化采样一致性算法边缘化外点去除误匹配。经实验对比,该方法相较于SURF准确性更高,实时性有较大提升,可满足多数复杂环境下的图像匹配。
