54 Pattern Recognition
نویسنده
چکیده
The two fundamental problems in a pattern recognition system are feature extraction (shape measurement) and classification. The problem of extracting a vector of shape measurements from a digital image can be further decomposed into three subproblems. The first is the image segmentation problem, i.e., the separation of objects of interest from their background. The cluster analysis methods discussed in Section 54.1 are useful here. The second subproblem is that of finding the objects in the segmented image. An example is the location of text lines in a document as illustrated in Section 54.2. The final subproblem is extracting the shape information from the objects detected. Here there are many tools available depending on the properties of the objects that are to be classified. The Hough transform (Section 54.2), polygonal approximation (Section 54.3), shape measurement (Section 54.4), and polygon decomposition (Section 54.6), are some of the favorite tools used here. Important to many of these tasks is finding a nice viewpoint from which extraction is robust and efficient (Section 54.5). Proximity graphs, discussed in Section 54.2, are used extensively for both cluster analysis and shape measurement. The classification problem involves the design of efficient decision rules with which to classify the feature vector. The most powerful decision rules are the nonparametric rules which make no assumptions about the underlying distributions of the feature vectors. Of these the nearest-neighbor (NN) rule, treated in Section 54.7, is the most well known. This section covers the three main issues concerning NN-rules: how to edit the data set so that little storage space is used, how to search for the nearest neighbor of a vector efficiently, and how to estimate the future performance of a rule both reliably and efficiently.
منابع مشابه
Local gradient pattern - A novel feature representation for facial expression recognition
Many researchers adopt Local Binary Pattern for pattern analysis. However, the long histogram created by Local Binary Pattern is not suitable for large-scale facial database. This paper presents a simple facial pattern descriptor for facial expression recognition. Local pattern is computed based on local gradient flow from one side to another side through the center pixel in a 3x3 pixels region...
متن کاملModeling of an Automatic CAD-Based Feature Recognition and Retrieval System for Group Technology Application
................................................................................................................................... 3 INTRODUCTION ......................................................................................................................... 11 Background .....................................................................................................................
متن کاملOn Creating Reference Templates for Speaker Independent Recognition of Isolated Words
The three aspects of a statistical approach to a pattern recognition problem are the selection of features, choice of a measure of similarity, and a method for creating the reference templates (patterns) used in the statistical tests. This paper discusses a philosophy for creating reference templates for a speaker independent, isolated word recognition system. Although there remain many unanswe...
متن کاملLocal Derivative Pattern with Smart Thresholding: Local Composition Derivative Pattern for Palmprint Matching
Palmprint recognition is a new biometrics system based on physiological characteristics of the palmprint, which includes rich, stable, and unique features such as lines, points, and texture. Texture is one of the most important features extracted from low resolution images. In this paper, a new local descriptor, Local Composition Derivative Pattern (LCDP) is proposed to extract smartly stronger...
متن کاملA Bayesian Approach for the Recognition of Control Chart Patterns
In this research, an iterative approach is employed to recognize and classify control chart patterns. To do this, by taking new observations on the quality characteristic under consideration, the Maximum Likelihood Estimator of pattern parameters is first obtained and then the probability of each pattern is determined. Then using Bayes’ rule, probabilities are updated recursively. Finally, when...
متن کامل