Search

This paper proposes an electronic part and component assembly system based on an industrial robot with binocular stereo vision. Firstly, binocular stereo vision with a visual attention mechanism model is used to get quickly the image regions which contain the electronic parts and components. Secondly, a deep neural network is adopted to recognize the features of the electronic parts and components. Thirdly, in order to control the end-effector of the industrial robot to grasp the electronic parts and components, a genetic algorithm (GA) is proposed to compute the transition matrix and the inverse kinematics of the industrial robot (end-effector), which plays a key role in bridging the binocular stereo vision and the industrial robot. Finally, the proposed assembly system is tested in LED component assembly experiments, and the results denote that it has high efficiency and good applicability.

In face recognition and classification, feature extraction and classification based on insufficient labeled data is a well-known challenging problem. In this paper, a novel semi-supervised learning algorithm named deep belief network embedded with Softmax regress (DBNESR) is proposed to address this problem. DBNESR first learns hierarchical representations of feature by deep learning and then makes more efficient classification with Softmax regress. At the same time we design many kinds of classifiers based on supervised learning: BP, HBPNNs, RBF, HRBFNNs, SVM and multiple classification decision fusion classifier——hybrid HBPNNs-HRBFNNs-SVM classifier. The conducted experiments validate: Firstly, the proposed semi-supervised deep learning algorithm DBNESR is optimal for face recognition with the highest and most stable recognition rates; Second, the semi-supervised learning algorithm has better effect than all supervised learning algorithms; Third, hybrid neural networks has better effect than single neural network.