Abstract
With the development of intelligent industrial production, industrial components with linear structure tend to be regular, such as TV LCD module, mobile phone screen, and electronic equipment shell. Recognition of linear structure objects by machine vision is an important aspect of intelligent industry. At present, shape matching algorithm is mostly used for arbitrary structure objects. It will be time-consuming if it is directly used to detect the linear structure objects as it needs to traverse the parameter space of the object. To solve the traversal problem and detect the linear structure objects in real time, a heuristic detection algorithm is designed according to the characteristics of linear structure objects. First, the coarse position and orientation are obtained by mean shift filtering and heuristic region grouping to reduce the searching range. Then, the heuristic search method is used to get the precise location information. The heuristic search method is designed based on the particle swarm optimization algorithm and heuristic information. The proposed method has been evaluated on two image databases of common industrial parts and backlight units which are typical linear structure objects. The experimental results showed that the proposed algorithm could reduce the detect time by more than 70% averagely while the detection accuracy is kept. It proves that the proposed algorithm can detect linear structure objects in real time and is suitable for the detection of objects with linear structures.
Highlights
It is necessary to use the vision system to detect these components to realize automatic assembly. e working principle of vision system is that it captures the images of the offline and online object and compares their differences to get the poses of the online object [5]. e offline object and its image are considered as the reference object and the reference image. e pose of the online object refers to the relative position and orientation compared to the offline object. en the pose is transformed into the robot coordinate system to perform specific works. erefore, it is necessary to detect the object and acquire its pose in the image fast and accurately
Mathematical Problems in Engineering categories. e general category refers to the methods that can be used in any applications, namely, they are suitable for objects with arbitrary shape, such as the shape matching method [6], modified generalized Hough transform (MGHT) [7], texture matching method [4], and deeplearning-based method [8, 9]. e features include edge, texture, color, or implied comprehensive characteristics. e shape matching method can detect the objects which have rigid shapes, such as the industrial parts
The detection methods for linear structure objects either directly use the general methods or use the modified methods, for example, a utility pole occlusion detection method based on the line segment detection algorithm [18], a fuzzy multiobjective approach based on a hybrid wavelet transform and fuzzy clustering method with multiobjective particle swarm optimization (PSO) for the Landsat images [19], and an improved feature extraction method based on LBP for bearing fault diagnosis [20]. ese methods are fast and effective, but they still need to traverse the parameter space which is time-consuming
Summary
Deep-learningbased methods can extract richer features which improve the detection accuracy greatly They are usually used in the case of natural scenes which can provide a large number of train images. The detection methods for linear structure objects either directly use the general methods or use the modified methods, for example, a utility pole occlusion detection method based on the line segment detection algorithm [18], a fuzzy multiobjective approach based on a hybrid wavelet transform and fuzzy clustering method with multiobjective particle swarm optimization (PSO) for the Landsat images [19], and an improved feature extraction method based on LBP for bearing fault diagnosis [20]. To solve the traversal problem and make the method applicable to general linear structure objects, this paper proposed a coarse-to-fine method which combines the heuristic information with mean shift segmentation and particle search optimization. To solve the traversal problem and make the method applicable to general linear structure objects, this paper proposed a coarse-to-fine method which combines the heuristic information with mean shift segmentation and particle search optimization. e contributions of this paper can be concluded as follows:
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
