文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > 高等教育 > 理学 > A Survey of Shape Feature Extraction Techniques
侵权投诉

A Survey of Shape Feature Extraction Techniques

A Survey of Shape Feature Extraction Techniques

44

Pattern Recognition Techniques, Technology and Applications occultation invariance: when some parts of a shape are occulted by other objects, the

feature of the remaining part must not change compared to the original shape.

statistically independent: two features must be statistically independent. This

represents compactness of the representation.

reliability: as long as one deals with the same pattern, the extracted features must

remain the same.

In general, shape descriptor is a set of numbers that are produced to represent a given shape feature. A descriptor attempts to quantify the shape in ways that agree with human intuition (or task-specific requirements). Good retrieval accuracy requires a shape descriptor to be able to effectively find perceptually similar shapes from a database. Usually, the descriptors are in the form of a vector. Shape descriptors should meet the following requirements:

the descriptors should be as complete as possible to represent the content of the

information items.

the descriptors should be represented and stored compactly. The size of a descriptor

vector must not be too large.

the computation of the similarity or the distance between descriptors should be simple;

otherwise the execution time would be too long.

Shape feature extraction and representation plays an important role in the following categories of applications:

shape retrieval: searching for all shapes in a typically large database of shapes that are

similar to a query shape. Usually all shapes within a given distance from the query are determined or the first few shapes that have the smallest distance.

shape recognition and classification: determining whether a given shape matches a

model sufficiently, or which of representative class is the most similar.

shape alignment and registration: transforming or translating one shape so that it best

matches another shape, in whole or in part.

shape approximation and simplification: constructing a shape with fewer elements

(points, segments, triangles, etc.), so that it is still similar to the original.

Many shape description and similarity measurement techniques have been developed in the past. A number of new techniques have been proposed in recent years. There are 3 main classification methods as follows:

contour-based methods and region-based methods [1]. This is the most common and

general classification and it is proposed by MPEG-7. It is based on the use of shape boundary points as opposed to shape interior points. Under each class, different methods are further divided into structural approaches and global approaches. This sub-class is based on whether the shape is represented as a whole or represented by segments/sections (primitives).

space domain and transform domain [2]. Methods in space domain match shapes on

point (or point feature) basis, while feature domain techniques match shapes on feature (vector) basis.

information preserving (IP) and non-information preserving (NIP). IP methods allow an

accurate reconstruction of a shape from its descriptor, while NIP methods are only capable of partial ambiguous reconstruction. For object recognition purpose, IP is not a requirement.

Unlike the traditional classification, the approaches of shape-based feature extraction and representation are classified according to their processing approaches: One-dimensional

上一页第2页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email