A general problem in automation is the correct supply of assembly parts to assembly units. At the same time it is of decisive importance that the stable orientation as well as eventual bad quality characteristics of the assembly parts can be recognized to prevent a faulty assembly, a standstill of machine, or, in the worst case, a destruction of the assembly unit. In practice, among mechanical devices, optoelectronic inspectautomates are used more
and more to solve this problem of correct sorting of assembly parts. These optoelectronic devices are checking the stable orientation and corresponding desired quality characteristics of assembly parts touch-less and automated. For solving this task it is necessary to calculate some features of the assembly part that has to be checked in the first step. The second step is to use these calculated features for a safe and stable classification of an assembly part. The calculation of the necessary features as well as the classification have to be realized in real-time. The amount of time that has to be kept, typically ranges from 5ms to 200ms depending on the assembly part that has to be checked and depending on the configuration of the optoelectronic inspectautomate. For that purpose features are calculated for a realized classification with which it is possible to check the stable orientation and corresponding desired quality characteristics for a multitude of assembly parts. But it is not possible to realize a safe and stable classification with these features, where the several stable orientations and with it the corresponding desired quality characteristics can only be classified
because of small details in the acquired images of the assembly part to be checked. In the cases of application where these described features cannot be used for a safe and stable classification it is necessary to introduce an additional feature with which it is possible to fulfill these applications with optoelectronic inspectautomates, because of its special characteristics. THerefore it is possible to increase the field of applications for such devices. This paper will introduce the feature Region of Interest which gives the possibility of using small details of an assembly part for a safe and stable classification of its stable orientation and quality characteristics. Additionally, this paper will describe the general implementation of the featuer Region of Interest for an optoelectronic inspectautomate.
Various methods for multivariate calibration like Principal Component Regression (PCR) and Partial Least Squares Regression (PLSR) are evaluated for their use in the field of pattern classification. These methods have the advantage that they can deal with high-dimensional feature spaces and multi-collinear data, since they inherently reduce the dimension of the feature space to represent it by one single dimension. Additionally, they yield very simple linear classifiers, which can be used for real-time calculation. These properties make the methods particularly useful in the field of image processing, where one often find high-dimensional spaces with linearly dependent data and usually we have tight requirements on computational complexity.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.