Three-Dimensional Geometric-Physical Modeling of an Environment with an In-House-Developed Multi-Sensor Robotic System DOI Creative Commons
Su Zhang, Meijuan Yu, Haoyu Chen

et al.

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(20), P. 3897 - 3897

Published: Oct. 20, 2024

Environment 3D modeling is critical for the development of future intelligent unmanned systems. This paper proposes a multi-sensor robotic system environmental geometric-physical and corresponding data processing methods. The primarily equipped with millimeter-wave cascaded radar multispectral camera to acquire electromagnetic characteristics material categories target environment simultaneously employs light detection ranging (LiDAR) an optical achieve three-dimensional spatial reconstruction environment. Specifically, sensor adopts multiple input output (MIMO) array obtains synthetic aperture images through 1D mechanical scanning perpendicular array, thereby capturing properties camera, nine channels, provides rich spectral information identification clustering. Additionally, LiDAR used obtain point cloud, combined RGB captured by enabling construction geometric model. By fusing from four sensors, comprehensive model can be constructed. Experiments conducted in indoor environments demonstrated excellent spatial-geometric-physical results. play important role various applications, such as planning.

Language: Английский

Enhancing the Performance of YOLOv9t Through a Knowledge Distillation Approach for Real-Time Detection of Bloomed Damask Roses in the Field DOI Creative Commons
Farhad Fatehi, Hossein Bagherpour, Jafar Amiri Parian

et al.

Smart Agricultural Technology, Journal Year: 2025, Volume and Issue: 10, P. 100794 - 100794

Published: Jan. 20, 2025

Language: Английский

Citations

1
Multi-Class Flower Counting Model with Zha-KNN Labelled Images Using Ma-Yolov9 DOI Open Access

A. Jasmine Xavier,

S. Valarmathy,

J. Gowrishankar

et al.

International Journal of Advanced Computer Science and Applications, Journal Year: 2024, Volume and Issue: 15(6)

Published: Jan. 1, 2024

The flowering period is a critical time for the growth of plants. Counting flowers can help farmers predict corresponding fields yield information. As there are several works proposed flower counting purposes, they lack prediction different with counts. Hence, novel model has been in this study. Initially, fed images as input, then these undergo pre-processing. In pre-processing, converted to grayscale improved accuracy, and noise removed using bilateral filters. Noise-removed given edge detection, GI-CED. Edge-detected augmented improve learning rate model. Augmented labeled ZHA-KNN. Labeled feature extracted MA-YoloV9, which pre-trained detect image count obtained output. Overall, was implemented an accuracy value about 98.8% F1-Score 92.2% far better than previous models.

Language: Английский

Citations

1
Three-Dimensional Geometric-Physical Modeling of an Environment with an In-House-Developed Multi-Sensor Robotic System DOI Creative Commons
Su Zhang, Meijuan Yu, Haoyu Chen

et al.

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(20), P. 3897 - 3897

Published: Oct. 20, 2024

Environment 3D modeling is critical for the development of future intelligent unmanned systems. This paper proposes a multi-sensor robotic system environmental geometric-physical and corresponding data processing methods. The primarily equipped with millimeter-wave cascaded radar multispectral camera to acquire electromagnetic characteristics material categories target environment simultaneously employs light detection ranging (LiDAR) an optical achieve three-dimensional spatial reconstruction environment. Specifically, sensor adopts multiple input output (MIMO) array obtains synthetic aperture images through 1D mechanical scanning perpendicular array, thereby capturing properties camera, nine channels, provides rich spectral information identification clustering. Additionally, LiDAR used obtain point cloud, combined RGB captured by enabling construction geometric model. By fusing from four sensors, comprehensive model can be constructed. Experiments conducted in indoor environments demonstrated excellent spatial-geometric-physical results. play important role various applications, such as planning.

Language: Английский

Citations

1