This paper studies the problem of detecting unknown objects within indoor environments in an active and natural manner. The visual saliency scheme utilizing both color and depth cues is proposed to arouse the interests of the machine system for detecting unknown objects at salient positions in a 3D scene. The 3D points at the salient positions are selected as seed points for generating object hypotheses using the 3D shape. We perform multi-class labeling on a Markov random field (MRF) over the voxels of the 3D scene, combining cues from object hypotheses and 3D shape. The results from MRF are further refined by merging the labeled objects, which are spatially connected and have high correlation between color histograms. Quantitative and qualitative evaluations on two benchmark RGB-D datasets illustrate the advantages of the proposed method. The experiments of object detection and manipulation performed on a mobile manipulator validate its effectiveness and practicability in robotic applications.

Robotic systems are usually controlled to repetitively perform specific actions for manufacturing tasks. The traditional control methods are domain-dependent and model-dependent with cost of much human efforts. They cannot meet the new requirements of generality and flexibility in many areas such as intelligent manufacturing and customized production. This paper develops a general model-free approach to enable robots to perform multi-step object sorting tasks through deep reinforcement learning. Taking projected heightmap images from different time steps as input without extra high-level image analysis and understanding, critic models are designed to produce a pixel-wise Q value map for each type of action. It is a new trial to apply pixel-wise Q value-based critic networks to solve multi-step sorting tasks that involve many types of actions and complex action constraints. The experimental validations on simulated and realistic object sorting tasks demonstrate the effectiveness of the proposed approach. Qualitative results (videos), code for simulated and realistic experiments, and pre-trained models are available at https://github.com/JiatongBao/DRLSorting

The application of remote digital video surveillance and image recognition technology in online monitoring of power equipment is conducive to timely equipment maintenance and troubleshooting. In order to solve the problem of slow speed and large amount of computation of traditional template matching algorithm for power image recognition, a second template matching algorithm for fast recognition of target image is proposed in this article. Firstly, a quarter of the template data is taken and matched within a quarter of the source image, and a reasonable error threshold is given in the matching process. Then, the neighborhood of the minimum error point in rough matching is matched to get the final result. Finally, the algorithm is applied to identify the power equipment and detect the abnormal state of the power equipment. The experimental results show that the matching algorithm can not only accurately locate and identify power equipment and detect equipment faults, but also greatly improve the matching speed compared with other commonly used template matching algorithms.