File(s) under permanent embargo
A Novel Bilinear Feature and Multi-Layer Fused Convolutional Neural Network for Tactile Shape Recognition
journal contributionposted on 2023-07-12, 02:17 authored by Jie Chu, Jueping Cai, He Song, Yuxin ZhangYuxin Zhang, Linyu Wei
Convolutional neural networks (CNNs) can automatically learn features from pressure information, and some studies have applied CNNs for tactile shape recognition. However, the limited density of the sensor and its flexibility requirement lead the obtained tactile images to have a low-resolution and blurred. To address this issue, we propose a bilinear feature and multi-layer fused convolutional neural network (BMF-CNN). The bilinear calculation of the feature improves the feature extraction capability of the network. Meanwhile, the multi-layer fusion strategy exploits the complementarity of different layers to enhance the feature utilization efficiency. To validate the proposed method, a 26 class letter-shape tactile image dataset with complex edges was constructed. The BMF-CNN model achieved a 98.64% average accuracy of tactile shape. The results show that BMF-CNN can deal with tactile shapes more effectively than traditional CNN and artificial feature methods.
bilinear featureChemistryChemistry, Analyticalconvolutional neural networkEngineeringEngineering, Electrical & ElectronicInstruments & Instrumentationmulti-layer fusionOBJECT RECOGNITIONPhysical SciencesScience & Technologytactile shapeTechnology3103 Ecology4008 Electrical engineering4009 Electronics, sensors and digital hardware4104 Environmental management4606 Distributed computing and systems softwareEcologyEnvironmental Science and Management not elsewhere classifiedElectrical and Electronic Engineering not elsewhere classifiedDistributed ComputingAnalytical Chemistry not elsewhere classified