WISEKO: a knowledge-based system for the recognition of 2D-complex objects in CAQ applications

Abstract

The automatical recognition of complex objects ( objects with one or more drills) represents a central scope of duties in the automatization of numerous production processes. Contrary to simple objects ( objects without drills) which outer contour form represents the single recognition feature, for complex objects further features must be determined and connected with each other to describe these objects.For the recognition of workpieces for example additional to the outer contour form of the workpiece, the number of drills, their forms as well as their normalized sizes must be determined and connected with each other. For industrial application in the field of automatic quality control ( CAQ) and Robotics the systems for the recognition of complex objects have to fulfil the requirement of flexibility and realtime processing. Relative to this subject the literature on image processing shows only a few concepts [4 ] ,[5 ] ,[6 ] ,[8 ], [12 ] . The reason for it is the lack of suitable preprocessing methods for feature extraction, especially connected with the contour extraction and its description. Based on the object-oriented method for contour extraction ( OCE) [13 ] , in this work the knowledge-based system for the recognition of complex objects called.WISEKO will be introduced. The system is built up in form of an expert system and enables the recognition of any complex object. The WISEKO-system is flexible and can be used for various applications in CAQ and Robotics. Because of using a hardware implementation for the OCE-method the WISEKO-system can be imployed for realtime-applications.

Content Based Image Retrieval Using Fuzzy Texton and Shearlet Transform

“…The first method describes the object outlines and the second describes the object areas. Suitable contour and region primitives are first used in order to extract the relevant objects from a given image (e.g., Freeman codes [18], polygon lines [19], and pixels or image squares [20]). Based on this data, shape description can be performed by applying a feature vector in order to represent the given objects.…”

“…Precision is given by (18) Recall is given by (19) Accuracy (20) Where is the number of images retrieved in the top N positions that are similar to the query image and M is the total number of images in the database similar to the query image. Table 1 shows the comparison results of Proposed Method with Texels of Fuzzy Textons (TFT) [21] and TT [29].…”

Object Shape Recognition Using Wavelet Descriptors

“…To derive an angle function, we use the polygon information of a given object shape derived after contour extraction and approximation [10]. Figure 1 shows the example of an object shape with five edges; its derived angle (green colored) and the periodical angle function * (blue colored) are shown in Figure 2.…”

“…In recent years, many methods for shape representation and recognition have been proposed [1][2][3][4][5][6][7][8][9][10]. An advanced review of shape representation techniques can be found in [1,2].…”

“…The first method describes the object outlines and the second describes the object areas. In both cases, suitable contour and region primitives are first used in order to extract the relevant objects from a given image (e.g., Freeman codes [3], polygon lines [10], and pixels or image squares [4]). Based on this data, shape description can be performed by applying a feature vector in order to represent the given objects.…”