Automated Surface Crack Detection in Historical Constructions with Various Materials Using Deep Learning-Based YOLO Network

International Journal of Architectural Heritage, Год журнала: 2024, Номер unknown, С. 1 - 17

Опубликована: Июль 17, 2024

Язык: Английский

---

Close Proximity Time-to-collision Prediction for Autonomous Robot Navigation: An Exponential GPR Approach

Alexandria Engineering Journal, Год журнала: 2022, Номер 61(12), С. 11171 - 11183

Опубликована: Май 9, 2022

Fusion of X-band Doppler radar and infrared sensors can offer a great advantage for close proximity time-to-collision (TTC) prediction in the field autonomous robot navigation due to precise obstacle speed detection direction sensing. Nevertheless, poor ranging performance from may result owing fluctuating reflectivity moving obstacle. The TTC accuracy also be further degraded when obstacle's trajectory is non-parallel robot's heading uncertainty radar's radiation pattern which typically increases at side lobes antenna. Thus, enhance performance, we propose an exponential Gaussian Process Regression (E-GPR)-based model able approximate unknown function probabilistic manner. proposed method validated via series experiments with approaching different viewing angles between 30 63 cm/s. Results demonstrate that average error based on sensor fusion 0.31s; but E-GPR method, successfully reduced 0.0937s, largest reduction compared against other competing machine-learning models such as multilayer perceptron neural network, support vector machine boosted tree.

Язык: Английский

---

Multi-Objective Location and Mapping Based on Deep Learning and Visual Slam

Sensors, Год журнала: 2022, Номер 22(19), С. 7576 - 7576

Опубликована: Окт. 6, 2022

Simultaneous localization and mapping (SLAM) technology can be used to locate build maps in unknown environments, but the constructed often suffer from poor readability interactivity, primary secondary information map cannot accurately grasped. For intelligent robots interact meaningful ways with their environment, they must understand both geometric semantic properties of scene surrounding them. Our proposed method not only reduce absolute positional errors (APE) improve positioning performance system also construct object-oriented dense point cloud output model each object reconstruct indoor scene. In fact, eight categories objects are for detection using coco weights our experiments, most actual reconstructed theory. Experiments show that number points is significantly reduced. The average error Technical University Munich (TUM) datasets very small. camera reduced introduction constraints, improved. At same time, algorithm segment environment high accuracy.

Язык: Английский

---

Using deep learning in an embedded system for real-time target detection based on images from an unmanned aerial vehicle: vehicle detection as a case study

International Journal of Digital Earth, Год журнала: 2023, Номер 16(1), С. 910 - 936

Опубликована: Март 14, 2023

For a majority of remote sensing applications unmanned aerial vehicles (UAVs), the data need to be downloaded ground devices for processing, but this procedure cannot satisfy demands real-time target detection. Our objective in study is develop system based on an embedded technology image acquisition, detection, transmission and display results, user interaction while providing support interactions between multiple UAVs users. This work divided into three parts: (1) We design technical framework implementation detection according application requirements. (2) efficient reliable module realize cross-platform communication airborne ground-side servers. (3) optimize YOLOv4 algorithm by using K-Means TensorRT inference improve accuracy speed NVIDIA Jetson TX2. In experiments involving static it had overall confidence 89.6% rate missed 3.8%; dynamic 88.2% 4.6%, respectively.

Язык: Английский

---

Continuous dynamic gesture recognition using surface EMG signals based on blockchain-enabled internet of medical things

Information Sciences, Год журнала: 2023, Номер 646, С. 119409 - 119409

Опубликована: Июль 22, 2023

Язык: Английский

---