Classifying Body and Surface Reflections Using Expectation-Maximization

نویسندگان

  • Hans Jørgen Andersen
  • Moritz Störring
چکیده

This paper presents a new method for the classification of dielectrical object’s RGB values into their body and surface reflections. Instead of segmenting into the two reflection components a weight is estimated that a given pixel belongs to one of them. A weighting value may be useful for classification of body and surface reflections in combination with other methods. The method operates globally on the pixel points using expectation maximization for fitting the body and surface vectors in the case of one highlight reflection. In the case of multiple highlights it is shown that it is possible to relax the method by fitting one surface vector to multiple highlights. The method was empirically validated on real image data captured using a high dynamic imaging sensor (120dB). Promising results show that the method is capable of classifying the two reflection components.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

The indexing ambiguity in serial femtosecond crystallography (SFX) resolved using an expectation maximization algorithm

Crystallographic auto-indexing algorithms provide crystal orientations and unit-cell parameters and assign Miller indices based on the geometric relations between the Bragg peaks observed in diffraction patterns. However, if the Bravais symmetry is higher than the space-group symmetry, there will be multiple indexing options that are geometrically equivalent, and hence many ways to merge diffra...

متن کامل

Named Entity Learning and Verification: Expectation Maximization in Large Corpora

The regularity of named entities is used to learn names and to extract named entities. Having only a few name elements and a set of patterns the a lgorithm learns new names and its elements. A verification step assures quality using a large background corpus. Further improvement is reached through classifying the newly learnt elements on character level. Moreover, unsupervised rule learning is ...

متن کامل

Automatic Tissue Classification for the Human Head from Multispectral MRI

A robust, automatic algorithm is presented for classifying multispectral MR scans of the human head into nine tissue classes. User initialization is not required. The proposed algorithm is based on three principles: (i) a combination of scalar and multispectral Expectation-Maximization clustering stages are used for fast, unsupervised statistical analysis, (ii) the minimal required amount of pr...

متن کامل

Tracking People in Presence of Occlusion

Tracking humans is a diicult problem because of the non-rigid nature of the human body, and the frequent occlusion encountered among people. We present a framework to track multiple people in a xed camera situation. Our framework deals implicitly with occlusion, and is able to correctly label people during occlusion. We rst segment a person into classes of similar color using the Expectation Ma...

متن کامل

Document Classification Using a Finite Mixture Model

We propose a new method of classifying documents into categories. We define for each category a finite mixture model based on soft clustering of words. We treat the problem of classifying documents as that of conducting statistical hypothesis testing over finite mixture models, and employ the EM algorithm to efficiently estimate parameters in a finite mixture model. Experimental results indicat...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2003