Possibilistic C-Means in Scene Matching
نویسندگان
چکیده
Determining if two images acquired at different times and under different viewing conditions contain the same scene is a difficult problem in computer vision. We demonstrate an approach that utilizes spatial relationships among the objects in the two scenes that ultimately produces a mapping of objects from one view to the other, and as a bonus, recovers the viewing transformation parameters. The core of the system relies on capturing spatial relationship information through Force Histograms, affine-invariant image descriptors. Object mapping across images is performed by finding the best correspondence map (FMAP) between force histograms in the two images. The major problem is that the number of potential FMAPS is huge, even for modest numbers of scene objects. Hence, some optimization is required. Similar feature vectors are observed from F-histogram matching defined in the best correspondence map. Therefore, dense regions in the feature space are suspected to contain these vectors. Possibilistic C-Means clustering (PCM) is used to find these dense regions. The centroids of these dense regions are used to generate an FMAP using a nearest-neighbor like approach. The best FMAP is selected and translated into an object map identifying the correspondences between objects in the two images.
منابع مشابه
طبقه‌بندی نظارت‌نشده تصاویر فراطیفی با الگوریتم خوشه‌بندی Possibilistic Fuzzy c-Means
روش‌های طبقه‌بندی از مهم‌ترین روش‌های استخراج اطلاعات از تصاویر سنجش از دوری می‌باشند که به طور مرسوم به دو دسته نظارت‌شده و نظارت‌نشده تقسیم می‌شوند. روش‌های نظارت‌شده نیازمند جمع‌آوری داده‌های آموزشی بوده و مستلزم صرف هزینه و زمان می‌باشند. در مقابل، روش‌های نظارت‌نشده فقط متکی بر داده‌های تصویری بوده و اغلب به صورت اتوماتیک انجام می‌شوند. روش‌های نظارت‌نشده نسبت به روش‌های نظارت‌شده اگر چه م...
متن کاملColor scene transform between images using Rosenfeld-Kak histogram matching method
In digital color imaging, it is of interest to transform the color scene of an image to the other. Some attempts have been done in this case using, for example, lαβ color space, principal component analysis and recently histogram rescaling method. In this research, a novel method is proposed based on the Resenfeld and Kak histogram matching algorithm. It is suggested that to transform the color...
متن کاملPattern Recognition of Strong Graphs Based on Possibilistic c-means and k-formulae Matching
متن کامل
Transactions on Engineering and Sciences, Vol. I, August 2013
This paper presents a latest survey of different technologies using fuzzy clustering algorithms. Clustering approach is widely used in biomedical field like image segmentation. A different methods are used for medical image segmentation like Improved Fuzzy C Means(IFCM), Possibilistic C Means(PCM),Fuzzy Possibilistic C Means(FPCM), Modified Fuzzy Possibilistic C Means(MFPCM) and Possibilistic F...
متن کاملBilateral Weighted Fuzzy C-Means Clustering
Nowadays, the Fuzzy C-Means method has become one of the most popular clustering methods based on minimization of a criterion function. However, the performance of this clustering algorithm may be significantly degraded in the presence of noise. This paper presents a robust clustering algorithm called Bilateral Weighted Fuzzy CMeans (BWFCM). We used a new objective function that uses some k...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2005