Segmentation and Location Computation of Bin Objects

Abstract

In this paper we present a stereo vision based system for segmentation and location computation of partially occluded objects in bin picking environments. Algorithms to segment partially occluded objects and to find the object location [midpoint,x, y and z co-ordinates] with respect to the bin area are proposed. The z co-ordinate is computed using stereo images and neural networks. The proposed algorithms is tested using two neural network architectures namely the Radial Basis Function nets and Simple Feedforward nets. The training results fo feedforward nets are found to be more suitable for the current application. The proposed stereo vision system is interfaced with an Adept SCARA Robot to perform bin picking operations. The vision system is found to be effective for partially occluded objects, in the absence of albedo effects. The results are validated through real time bin picking experiments on the Adept Robot.

Deep Workpiece Region Segmentation for Bin Picking

“…Especially the usage of fully convolutional networks seems promising [13], [14] which is why our framework adapts the same general structure. Some segmentation methods are also specifically configured for bin picking tasks like [15], [16], [17]. Other than these methods, our framework considers only object-background segmentation as part of semantic segmentation, i.e., only two classes should be distinguished, namely important objects to interact with in the scene and background.…”

Robot Vision in Industrial Assembly and Quality Control Processes

“…On the other hand, a general so-called »Bin-picking«, where the objects have a 3-D shape and are randomly organized in a box, still remains a problem. Despite of many research works, that offer special solutions and improvements in the overcoming the bin-picking problem (Schraft & Ledermann, 2003;Kirkegaard, 2005;Kirkegaard & Moeslud, 2006;Hema et. al., 2007), the oldest challenge in robotics remains still unsolved.…”

“…Based on laser scanners for 3-D object detection a model based approach in combination with CAD models (Schraft & Ledermann, 2003;Kristensen et al, 2001) or Harmonic Shape Context features (Kirkegaard, 2005;Kirkegaard & Moeslund, 2006), which are invariant to translation, scale and 3-D rotation, have been applied. Also some research has been done applying stereo vision together with a set of two neural networks (which are then compared with each other) namely the Radial Basis Function nets and Simple Feed forward nets (Hema et al, 2007). An algorithm for segmentation of partially occluded bin objects and the location of the topmost object is proposed.…”

Vision based Systems for Localization in Service Robots

“…Fig.3 shows images samples of the added images and the distance of the obstacle images with respect to the stereo sensors. The features extracted from the added images are found to be good candidates for distance computations using neural networks [Hema et al, 2007]. The x, y and z co-ordinate information determined from the stereo images can be effectively used to locate obstacles and signs which can aid in collision free navigation in an indoor environment.…”

“…Using the reference image, the x and y co-ordinates is computed by finding the centroid of the obstacle image. The z co-ordinate can be computed using the unified stereo imaging principle proposed in [Hema et al, 2007]. The unified stereo imaging principle uses a morphological 'add' operation to add the left and right images acquired at a given distance.…”