Integrated edge and junction detection with the boundary tensor

Kothe,

doi:10.1109/iccv.2003.1238377

Cited by 43 publications

(28 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To this end, we present first detailed recordings of SHG signals from muscle slices of normal (wild-type, wt) and dystrophic mdx mice at different ages to emphasize the progression of ultrastructural deviation with age. For quantitative morphometric analysis of the 3-D aspect of tissue slices, we apply image processing algorithms based on the boundary tensor [14] (see Section III-A).…”

mentioning

confidence: 99%

Automated Multiscale Morphometry of Muscle Disease From Second Harmonic Generation Microscopy Using Tensor-Based Image Processing

Garbe

Buttgereit

Schürmann

et al. 2012

IEEE Trans. Biomed. Eng.

View full text Add to dashboard Cite

Abstract-Practically, all chronic diseases are characterized by tissue remodeling that alters organ and cellular function through changes to normal organ architecture. Some morphometric alterations become irreversible and account for disease progression even on cellular levels. Early diagnostics to categorize tissue alterations, as well as monitoring progression or remission of disturbed cytoarchitecture upon treatment in the same individual, are a new emerging field. They strongly challenge spatial resolution and require advanced imaging techniques and strategies for detecting morphological changes. We use a combined second harmonic generation (SHG) microscopy and automated image processing approach to quantify morphology in an animal model of inherited Duchenne muscular dystrophy (mdx mouse) with age. Multiphoton XYZ image stacks from tissue slices reveal vast morphological deviation in muscles from old mdx mice at different scales of cytoskeleton architecture: cell calibers are irregular, myofibrils within cells are twisted, and sarcomere lattice disruptions (detected as "verniers") are larger in number compared to samples from healthy mice. In young mdx mice, such alterations are only minor. The boundarytensor approach, adapted and optimized for SHG data, is a suitable approach to allow quick quantitative morphometry in whole tissue slices. The overall detection performance of the automated algorithm compares very well with manual "by eye" detection, the latter being time consuming and prone to subjective errors. Our algorithm outperfoms manual detection by time with similar reliability. This approach will be an important prerequisite for the implementation of a clinical image databases to diagnose and monitor specific morphological alterations in chronic (muscle) diseases.

show abstract

mentioning

confidence: 99%

Automated Multiscale Morphometry of Muscle Disease From Second Harmonic Generation Microscopy Using Tensor-Based Image Processing

Garbe

Buttgereit

Schürmann

et al. 2012

IEEE Trans. Biomed. Eng.

View full text Add to dashboard Cite

show abstract

“…文献中已提出许多二值化方法,本文采用 Niblack [25] 的自适应二值化算法(Wolf [26] 认为它是最好的二值化算法),该算法利用局部均值和标准差自动地为某一位置选取合适的二值化阈值: (3))类似.但考虑到距离角点太近的像素点的梯度方向不太稳定,为了减小它们对拟合角点位置时所产生不稳定性,应分配较小的权重.因此,支撑像素点到角点初始检测位置的距离的加权不再使用函数 ,而使用文献 [22,23] 中使用的权重函数: …”

Section: 确定支撑像素unclassified

Corner Detection and Sub-Pixel Localization Based on Local Orientation Distribution

Wang¹,

Wu²,

Wang³

2009

Journal of Software

View full text Add to dashboard Cite

detector etc, usually perform poor in accuracy of localization. In this paper, a novel algorithm for corner detection and localization is proposed, which is based on local orientation distribution (LOD). The algorithm consists of two steps: The first step is to detect corners by using the absolute corner energy and the relative corner energy. Then, the detected corners are re-located to sub-pixel accuracy by fitting the intersection of orientation lines of support pixels by the least squares technique. Experiments show that, the LOD-based algorithm can not only provides a higher localization accuracy than most popular detectors, but also has a better performance in resisting noise.

show abstract

“…Förstner and Gülch [12] propose a two-step procedure for localizing interest points. The first points are detected by searching for optimal windows using the auto-correlation matrix A.…”

Section: Interest Point Detectionmentioning

confidence: 99%

“…Köthe [12] improved the structure tensor computation using an increased resolution and non-linear averaging to optimize the localization accuracy. In our implementation, we applied a modified version of a Förstner corner detector that embraces some extensions proposed by Köthe.…”

Section: Interest Point Detectionmentioning

confidence: 99%

Text Detection in Scene Images Based on Interest Points

Nguyen¹,

Lee²

2015

J Inf Process Syst

View full text Add to dashboard Cite

Text in images is one of the most important cues for understanding a scene. In this paper, we propose a novel approach based on interest points to localize text in natural scene images. The main ideas of this approach are as follows: first we used interest point detection techniques, which extract the corner points of characters and center points of edge connected components, to select candidate regions. Second, these candidate regions were verified by using tensor voting, which is capable of extracting perceptual structures from noisy data. Finally, area, orientation, and aspect ratio were used to filter out non-text regions. The proposed method was tested on the ICDAR 2003 dataset and images of wine labels. The experiment results show the validity of this approach.

show abstract

Integrated edge and junction detection with the boundary tensor

Cited by 43 publications

References 13 publications

Automated Multiscale Morphometry of Muscle Disease From Second Harmonic Generation Microscopy Using Tensor-Based Image Processing

Automated Multiscale Morphometry of Muscle Disease From Second Harmonic Generation Microscopy Using Tensor-Based Image Processing

Corner Detection and Sub-Pixel Localization Based on Local Orientation Distribution

Text Detection in Scene Images Based on Interest Points

Contact Info

Product

Resources

About