A reinforced CenterNet scheme on position detection of acoustic levitated objects

• Published in: Neural computing & applications 2023-04, Vol.35 (12), p.8987-9002
• Main Authors: Li, Xinbo, Wang, Yingwei, Jiang, Liangxu, Chen, Ziyi, Fan, Shuyuan
• Format: Article
• Language: English
• Subjects: Accuracy; Acoustics; Algorithms; Artificial Intelligence; Artificial neural networks; Cameras; Computational Biology/Bioinformatics; Computational Science and Engineering; Computer Science; Data Mining and Knowledge Discovery; Datasets; Field programmable gate arrays; Image Processing and Computer Vision; Light emitting diodes; Methods; Modules; Neural networks; Original Article; Probability and Statistics in Computer Science; Representations
• ISSN: 0941-0643; 1433-3058
• DOI: 10.1007/s00521-022-08140-1

Position detection is essential for precise contactless manipulation as it can play an important role in analyzing the behavior patterns and motion regularities of levitated objects. However, traditional detection methods have several limitations, such as finite feature representation, low detection accuracy, and poor adaptability, notably for small targets. To address these issues, this paper proposes an effective detector called RFRM-CenterNet, which is the first attempt to detect the location of levitated objects based on a convolutional neural network. First, a receptive field refinement module consisting of multi-stage parallel dilated convolutional layers is designed to detect small levitated objects. Then, to increase the feature representation capability, a feature fusion network is designed, which employs a parallel fusion of ResNet50 and a receptive field refinement module. In addition, an experimental system is constructed, and a dataset is established to train the model and verify the performance of the proposed method. The experimental results show that RFRM-CenterNet has better performance in detecting levitated objects than other detection methods.