Object matching between two-dimensional images is an important problem in computer vision. The purpose of object matching is to decide the similarity between two objects. A new robust image matching method based on di...Object matching between two-dimensional images is an important problem in computer vision. The purpose of object matching is to decide the similarity between two objects. A new robust image matching method based on distance reciprocal was presented. The distance reciprocal is based on human visual perception. This method is simple and effective. Moreover, it is robust against noise. The experiments show that this method outperforms the Hausdorff distance, when the images with noise interfered need to be recognized.展开更多
The paper analyze and improve the SIFT optimized algorithm, and proposes an image matching method for SIFT algorithm based on quasi Euclidean distance and KD-tree. Experiments show that this algorithm has matching mor...The paper analyze and improve the SIFT optimized algorithm, and proposes an image matching method for SIFT algorithm based on quasi Euclidean distance and KD-tree. Experiments show that this algorithm has matching more points, high matching accuracy, no repealed points and higher advantage of matching efficiency based on keeping the basic characteristics of SIFT algorithm unchanged, and provides precise matching point to generate precise image stitching and other related fields of the follow-up product. At the same time, this method was applied to the layout optimization and achieved good results.展开更多
The mean Hausdorff distance, though highly applicable in image registration, does not work well on partial matching images. An improvement upon traditional Hausdorff-distance-based image registration method is propose...The mean Hausdorff distance, though highly applicable in image registration, does not work well on partial matching images. An improvement upon traditional Hausdorff-distance-based image registration method is proposed, which consists of the following two aspects. One is to estimate transformation parameters between two images from the distributions of geometric property differences instead of establishing explicit feature correspondences. This procedure is treated as the pre-registration. The other aspect is that mean Hausdorff distance computation is replaced with the analysis of the second difference of generalized Hausdorff distance so as to eliminate the redundant points. Experimental results show that our registration method outperforms the method based on mean Hausdorff distance. The registration errors are noticeably reduced in the partial matching images.展开更多
This paper describes target detection improvement in a distance measurement system using two rotatable cameras for floating production, storage and offloading （FPSO） facilities. The authors have developed a distance...This paper describes target detection improvement in a distance measurement system using two rotatable cameras for floating production, storage and offloading （FPSO） facilities. The authors have developed a distance measurement system that consists of two rotatable cameras on a ship and a target on a wharf for the automatic berthing of ships. This system measures a distance by detecting and tracking the target on the wharf using the two rotatable cameras on the ship. Our goal is to apply this distance measurement system to an automatic relative positioning system for a ship at an FPSO facility. In this application, the shape of the target in the images captured by the cameras is deformed by their relative positions and attitudes, which increases the measurement errors. To solve this problem, we propose a target detection method that improves the target deformations. The proposed target detection method is able to detect the deformed targets using a target database that is created by image conversion with the perspective projection of a reference target. By using the proposed target detection method, the distance measurement error is decreased. Experimental results on a miniature scale and in an indoor environment confirmed that the measurement error of the relative distance is decreased by using the proposed target detection method.展开更多
对2幅不同角度、不同光照条件或不同相机采集到的图像进行配准,是一项十分具有挑战性的研究。针对参考图像和待配准图像对之间存在的仿射变换问题,提出了一种灵活通用的、基于SIFT特征和角度相对距离的图像配准算法。算法充分利用了图...对2幅不同角度、不同光照条件或不同相机采集到的图像进行配准,是一项十分具有挑战性的研究。针对参考图像和待配准图像对之间存在的仿射变换问题,提出了一种灵活通用的、基于SIFT特征和角度相对距离的图像配准算法。算法充分利用了图像正确匹配特征点对之间存在的角度关系,实现了特征点之间的精确匹配。将所提算法同LLT(locally linear transforming)算法及RANSAC算法进行了对比实验,结果表明,新算法有较高的有效性和鲁棒性。而且新算法不仅适用于普通图像,在近红外与可见光图像以及遥感图像中均充分体现了良好的鲁棒性和适用性,在匹配特征点数目较少时,也具有良好的鲁棒性。展开更多
文摘Object matching between two-dimensional images is an important problem in computer vision. The purpose of object matching is to decide the similarity between two objects. A new robust image matching method based on distance reciprocal was presented. The distance reciprocal is based on human visual perception. This method is simple and effective. Moreover, it is robust against noise. The experiments show that this method outperforms the Hausdorff distance, when the images with noise interfered need to be recognized.
文摘The paper analyze and improve the SIFT optimized algorithm, and proposes an image matching method for SIFT algorithm based on quasi Euclidean distance and KD-tree. Experiments show that this algorithm has matching more points, high matching accuracy, no repealed points and higher advantage of matching efficiency based on keeping the basic characteristics of SIFT algorithm unchanged, and provides precise matching point to generate precise image stitching and other related fields of the follow-up product. At the same time, this method was applied to the layout optimization and achieved good results.
基金Project(61070090)supported by the National Natural Science Foundation of ChinaProject(2012J4300030)supported by the GuangzhouScience and Technology Support Key Projects,China
文摘The mean Hausdorff distance, though highly applicable in image registration, does not work well on partial matching images. An improvement upon traditional Hausdorff-distance-based image registration method is proposed, which consists of the following two aspects. One is to estimate transformation parameters between two images from the distributions of geometric property differences instead of establishing explicit feature correspondences. This procedure is treated as the pre-registration. The other aspect is that mean Hausdorff distance computation is replaced with the analysis of the second difference of generalized Hausdorff distance so as to eliminate the redundant points. Experimental results show that our registration method outperforms the method based on mean Hausdorff distance. The registration errors are noticeably reduced in the partial matching images.
文摘This paper describes target detection improvement in a distance measurement system using two rotatable cameras for floating production, storage and offloading （FPSO） facilities. The authors have developed a distance measurement system that consists of two rotatable cameras on a ship and a target on a wharf for the automatic berthing of ships. This system measures a distance by detecting and tracking the target on the wharf using the two rotatable cameras on the ship. Our goal is to apply this distance measurement system to an automatic relative positioning system for a ship at an FPSO facility. In this application, the shape of the target in the images captured by the cameras is deformed by their relative positions and attitudes, which increases the measurement errors. To solve this problem, we propose a target detection method that improves the target deformations. The proposed target detection method is able to detect the deformed targets using a target database that is created by image conversion with the perspective projection of a reference target. By using the proposed target detection method, the distance measurement error is decreased. Experimental results on a miniature scale and in an indoor environment confirmed that the measurement error of the relative distance is decreased by using the proposed target detection method.
文摘对2幅不同角度、不同光照条件或不同相机采集到的图像进行配准,是一项十分具有挑战性的研究。针对参考图像和待配准图像对之间存在的仿射变换问题,提出了一种灵活通用的、基于SIFT特征和角度相对距离的图像配准算法。算法充分利用了图像正确匹配特征点对之间存在的角度关系,实现了特征点之间的精确匹配。将所提算法同LLT(locally linear transforming)算法及RANSAC算法进行了对比实验,结果表明,新算法有较高的有效性和鲁棒性。而且新算法不仅适用于普通图像,在近红外与可见光图像以及遥感图像中均充分体现了良好的鲁棒性和适用性,在匹配特征点数目较少时,也具有良好的鲁棒性。
