Skip to main content
Fachgebiete
Automobil + Motoren Bauwesen + Immobilien Business IT + Informatik Elektrotechnik + Elektronik Energie + Nachhaltigkeit Finance + Banking Management + Führung Marketing + Vertrieb Maschinenbau + Werkstoffe Versicherung + Risiko
DE
EN
Bücher
Zeitschriften
Themenseiten
Organisationspsychologie Projektmanagement Marketing Smart Manufacturing
Weitere Formate
Podcasts Webinare Technik Webinare Wirtschaft Kongresse Veranstaltungskalender Awards
Jetzt Einzelzugang starten
Zugang für Unternehmen
Referenzkunden
SLX-Digitalkonferenz: Zukunftswerkstatt 2024

Mehr Umsatz mit Top-Kontakten

Am 7. Mai erfahren Sie, wie Vertriebsteams Leadgenerierung, Leadnurturing und Leadstrategien effizient steuern können.
Erweiterte Suche
Anmelden
nach oben

2016 | Buch

Introduction Kapitel lesen Erstes Kapitel lesen

Machine Vision

Automated Visual Inspection: Theory, Practice and Applications

verfasst von: Jürgen Beyerer, Fernando Puente León, Christian Frese

Verlag: Springer Berlin Heidelberg

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten
insite
SUCHEN

Über dieses Buch

The book offers a thorough introduction to machine vision. It is organized in two parts. The first part covers the image acquisition, which is the crucial component of most automated visual inspection systems. All important methods are described in great detail and are presented with a reasoned structure.
The second part deals with the modeling and processing of image signals and pays particular regard to methods, which are relevant for automated visual inspection.

Inhaltsverzeichnis

Frontmatter
1. Introduction
Abstract
Machine vision and automated visual inspection are rapidly finding their way into industrial measurement and quality control in the technical and engineering sectors. This development especially benefits from increasingly powerful computers and reasonably priced camera components.
This book provides insight into the fascinating and very up-to-date topic of automated visual inspection and image processing. An extensive content is presented in an easily comprehensible way and is explained using various examples. No particular previous knowledge is required.
Jürgen Beyerer, Fernando Puente León, Christian Frese

Image Acquisition

Frontmatter
2. Light
Abstract
Optical image acquisition and measurement methods rely on the properties of light. Light can be described physically in two very different ways: as a wave or as a particle. Depending on the concrete physical scenario in question, the one or the other perspective is more useful for describing an observed effect. Both of them are legitimate, but neither explains all observed phenomena. This is often referred to as the wave–particle duality.
Jürgen Beyerer, Fernando Puente León, Christian Frese
3. Optical Imaging
Abstract
Ideally, the imaging optics of an automated visual inspection system images every relevant point of the test object to exactly one point on the sensor. In other words, the diverging light bundle coming from a point G of the test object is transformed into a light bundle converging at some point B on the sensor. Such a correspondence of two points B = A{G} yields a sharp optical imaging.
Jürgen Beyerer, Fernando Puente León, Christian Frese
4. Radiometry
Abstract
The field of photometry is concerned with measuring the intensity and power of visible light with respect to the sensitivity of the human eye (see Chap. 5). Visible light is only a part of the whole electromagnetic spectrum (see Sec. 2.1.1). The respective wavelengths range from about 380 nm to 780 nm. Of course, physical measurements of intensity, power, etc., can be obtained for the whole electromagnetic spectrum without respect to the human eye, as is done in the domain of radiometry or radiation physics.
Jürgen Beyerer, Fernando Puente León, Christian Frese
5. Color
Abstract
In photometry, all quantities are with reference to the perception of brightness by the human eye. The sensitivity of the human eye can be described by a function V (λ) of the wavelength. This so-called luminosity function varies for different ambient conditions, such as photopic vision or scotopic vision, for example (Fig. 5.1), which refer to human perception at daylight or in darkness. The luminosity function of the light-adapted eye is used for defining the photometric base system. It is scaled to have a maximum value of 1. The luminosity function can be empirically measured using a psychophysical method similar to the method described in Sec. 5.2.3. However, the perception of brightness of the human eye is not a metric quantity: differences and ratios cannot be quantified by human perception. The corresponding conclusions for photometric quantities do not represent the human perception of brightness [36].
Jürgen Beyerer, Fernando Puente León, Christian Frese
6. Sensors for Image Acquisition
Abstract
The image of a scene that has to be inspected is imaged to the image plane by imaging optics. For further processing of the resulting image, the spatial distribution of the irradiance is first converted into an analog electrical signal, which is then spatially and temporally sampled, quantized, and finally stored.
Jürgen Beyerer, Fernando Puente León, Christian Frese
7. Methods of Image Acquisition
Abstract
This chapter covers different methods for image acquisition in automated visual inspection. Selecting the appropriate acquisition method depends on the properties of interest of the investigated object (Fig. 7.1):
  • optical properties like reflectance, color, texture, and index of refraction as a function of position, or
  • geometrical properties like the three-dimensional shape of the object.
Jürgen Beyerer, Fernando Puente León, Christian Frese

Image Processing

Frontmatter
8. Image Signals
Abstract
An image signal g(x) acquired by the methods described in Chap.7 is a function g : \({{\mathbb{R}}^{2}}\to {{\mathbb{R}}^{Q}}\), which—in the general case—maps the whole image plane to vectorial values, where Q denotes the number of channels (cf. Sec. 1.3). At first, both the domain and the range of the image signal are considered to be continuous. In this case, g(x) is called a continuous image signal or an analog image signal.
Jürgen Beyerer, Fernando Puente León, Christian Frese
9. Preprocessing and Image Enhancement
Abstract
The main aims of preprocessing and image enhancement are
  • to obtain visually informative images, as well as
  • to ease the subsequent signal processing and automated image evaluation steps.
The rather simple image enhancement techniques, which are covered in the following section, are mainly used for improving the visual impression of an image. Section 9.2 introduces methods which can reduce the influence of systematic perturbations caused by inhomogeneous illumination or by poor image acquisition, for example. Section 9.3 is devoted to the suppression of random noise by using linear and nonlinear filters and finally, Sec. 9.4 discusses the topic of image registration.
Jürgen Beyerer, Fernando Puente León, Christian Frese
10. Image Restoration
Abstract
The image enhancement methods covered in Chap. 9 mainly considered subjective or qualitative criteria. In particular, the visual interpretability of the results was of greater importance than the ‘conservation of the original image traits’. This motivation resulted in mostly heuristic methods, which even altered the images to form pseudo-color images or false-color images (Sec. 9.1.3).
Jürgen Beyerer, Fernando Puente León, Christian Frese
11. Segmentation
Abstract
Segmentation tries to decompose an image g(x) into separate, meaningful areas. For example, the test object is isolated from the background, or the borders between different objects are detected. An automated analysis of a segmented image is often easier than that of an unprocessed image. The result of the segmentation could be used to determine the position and orientation of the segmented objects in a subsequent processing step.
Jürgen Beyerer, Fernando Puente León, Christian Frese
12. Morphological Image Processing
Abstract
Morphological methods used in the algebra of sets can be used for morphological image processing. They were introduced by Matheron and Serra under the term ‘Mathematical Morphology’ [12, 16, 17]. Here, image signals are considered to be point sets and morphological filters are operations manipulating these sets.
Jürgen Beyerer, Fernando Puente León, Christian Frese
13. Texture Analysis
Abstract
A texture can be defined as a two-dimensional structure with a certain deterministic or statistical regularity. However, there is no standard definition for the term ‘texture’ [13]. Alternatively, a texture can be defined, e.g., by the variations of the image structures, which are smaller than the scales of interest [25].
Jürgen Beyerer, Fernando Puente León, Christian Frese
14. Detection
Abstract
The term detection refers to the recognition of known or unknown objects in an image and to the determination of their position and orientation. On the one hand, the objects that are to be detected can be test objects, whose presence, orientation or integrity has to be inspected. On the other hand, it might be necessary to detect defects or certain structures such as, e.g., features, in the image.
Jürgen Beyerer, Fernando Puente León, Christian Frese
15. Image Pyramids, the Wavelet Transfm and Multiresolution Analysis
Abstract
Multiresolution analysis is a mathematical concept describing signal analysis on different resolution levels. In contrast to the analysis of the fine details contained in the image, which requires the maximum possible resolution, coarse structures can be analyzed with little effort using a reduced resolution. Two methods for multiresolution analysis, which are covered in this chapter, are image pyramids and the wavelet transform.
Jürgen Beyerer, Fernando Puente León, Christian Frese

Appendix

Frontmatter
16. A Mathematical Foundations
Abstract
The intercept theorem [1,2] is an important principle for constructing the optical paths in the model of geometric optics (Chap. 3).
Jürgen Beyerer, Fernando Puente León, Christian Frese
Backmatter
Metadaten
Titel
Machine Vision
verfasst von
Jürgen Beyerer
Fernando Puente León
Christian Frese
Copyright-Jahr
2016
Verlag
Springer Berlin Heidelberg
Electronic ISBN
978-3-662-47794-6
Print ISBN
978-3-662-47793-9
DOI
https://doi.org/10.1007/978-3-662-47794-6

Neuer Inhalt

    Bildnachweise