Image Segmentation of Cabin Assembly Scene Based on Improved RGB-D Mask R-CNN

Fu, Yichen; Fan, Junfeng; Xing, Shiyu; Wang, Zhe; Jing, Fengshui; Tan, Min

doi:10.1109/tim.2022.3145388

Cited by 13 publications

(5 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Band combination is the simplest and commonly used method in multi-source remote sensing data fusion [6,27,29,66,67], and different combination methods showed different effects for different tree species. The identification accuracy of DGB and PCA-D was 75.5% and 76.2%, respectively, which indicates a good identification performance.…”

Section: Discussionmentioning

confidence: 99%

Individual Tree Species Identification for Complex Coniferous and Broad-Leaved Mixed Forests Based on Deep Learning Combined with UAV LiDAR Data and RGB Images

Zhong,

Zhang,

Liu

et al. 2024

Forests

View full text Add to dashboard Cite

Automatic and accurate individual tree species identification is essential for the realization of smart forestry. Although existing studies have used unmanned aerial vehicle (UAV) remote sensing data for individual tree species identification, the effects of different spatial resolutions and combining multi-source remote sensing data for automatic individual tree species identification using deep learning methods still require further exploration, especially in complex forest conditions. Therefore, this study proposed an improved YOLOv8 model for individual tree species identification using multisource remote sensing data under complex forest stand conditions. Firstly, the RGB and LiDAR data of natural coniferous and broad-leaved mixed forests under complex conditions in Northeast China were acquired via a UAV. Then, different spatial resolutions, scales, and band combinations of multisource remote sensing data were explored, based on the YOLOv8 model for tree species identification. Subsequently, the Attention Multi-level Fusion (AMF) Gather-and-Distribute (GD) YOLOv8 model was proposed, according to the characteristics of the multisource remote sensing forest data, in which the two branches of the AMF Net backbone were able to extract and fuse features from multisource remote sensing data sources separately. Meanwhile, the GD mechanism was introduced into the neck of the model, in order to fully utilize the extracted features of the main trunk and complete the identification of eight individual tree species in the study area. The results showed that the YOLOv8x model based on RGB images combined with current mainstream object detection algorithms achieved the highest mAP of 75.3%. When the spatial resolution was within 8 cm, the accuracy of individual tree species identification exhibited only a slight variation. However, the accuracy decreased significantly with the decrease of spatial resolution when the resolution was greater than 15 cm. The identification results of different YOLOv8 scales showed that x, l, and m scales could exhibit higher accuracy compared with other scales. The DGB and PCA-D band combinations were superior to other band combinations for individual tree identification, with mAP of 75.5% and 76.2%, respectively. The proposed AMF GD YOLOv8 model had a more significant improvement in tree species identification accuracy than a single remote sensing sources and band combinations data, with a mAP of 81.0%. The study results clarified the impact of spatial resolution on individual tree species identification and demonstrated the excellent performance of the proposed AMF GD YOLOv8 model in individual tree species identification, which provides a new solution and technical reference for forestry resource investigation combined multisource remote sensing data.

show abstract

Section: Discussionmentioning

confidence: 99%

Individual Tree Species Identification for Complex Coniferous and Broad-Leaved Mixed Forests Based on Deep Learning Combined with UAV LiDAR Data and RGB Images

Zhong,

Zhang,

Liu

et al. 2024

Forests

View full text Add to dashboard Cite

show abstract

“…While traditional segmentation methods primarily leverage threshold setting, histogram analysis [14], region growing, fuzzy clustering [15,16], K-means clustering [17], and edge detection [18,19], advanced techniques incorporate active contours, graph cuts, and sophisticated mathematical and probabilistic models [20]. Notably, deep learning approaches [21][22][23][24][25] like Fully Convolutional Networks (FCN) [26],U-Net [27],PSPNet [28] and FC-DenseNet have revolutionized segmentation with their high precision in pixel-level classification. Deep learning methods offer unmatched segmentation accuracy and efficiency; however, automated threshold segmentation techniques remain popular for their simplicity and effectiveness [29,30].…”

Section: Image Segmentationmentioning

confidence: 99%

Advancing image segmentation with DBO-Otsu: Addressing rubber tree diseases through enhanced threshold techniques

Xie,

Wu,

Tang

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

Addressing the profound impact of Tapping Panel Dryness (TPD) on yield and quality in the global rubber industry, this study introduces a cutting-edge Otsu threshold segmentation technique, enhanced by Dung Beetle Optimization (DBO-Otsu). This innovative approach optimizes the segmentation threshold combination by accelerating convergence and diversifying search methodologies. Following initial segmentation, TPD severity levels are meticulously assessed using morphological characteristics, enabling precise determination of optimal thresholds for final segmentation. The efficacy of DBO-Otsu is rigorously evaluated against mainstream benchmarks like Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), and Feature Similarity Index (FSIM), and compared with six contemporary swarm intelligence algorithms. The findings reveal that DBO-Otsu substantially surpasses its counterparts in image segmentation quality and processing speed. Further empirical analysis on a dataset comprising TPD cases from level 1 to 5 underscores the algorithm’s practical utility, achieving an impressive 80% accuracy in severity level identification and underscoring its potential for TPD image segmentation and recognition tasks.

show abstract

“…Most deep learning stair detection methods [ 4 , 7 , 8 ] focus on extracting stair features in monocular vision through a CNN, and there is no deep learning method to make full use of the complementary relationship between the RGB map and the depth map for stair detection. Regarding the RGB-D fusion methods for deep learning, some methods fuse features in the input and output locations by simple summation and concatenation [ 14 , 15 , 16 , 17 , 18 ], and some methods design special modules to explore the implicit relationship between the two modalities [ 19 , 20 , 21 , 22 ]. This section briefly introduces some RGB-D-based stair detection methods and some RGB-D fusion methods for deep learning.…”

Section: Related Workmentioning

confidence: 99%

“…Refs. [ 15 , 16 ] directly adjust the three-channel RGB input to the four-channel RGB-D input, and the depth information is sent to the network as the fourth input channel. Refs.…”

Section: Related Workmentioning

confidence: 99%

RGB-D-Based Stair Detection and Estimation Using Deep Learning

Wang

Pei

Qiu

et al. 2023

Sensors

View full text Add to dashboard Cite

Stairs are common vertical traffic structures in buildings, and stair detection tasks are important in environmental perception for autonomous mobile robots. Most existing algorithms have difficulty combining the visual information from binocular sensors effectively and ensuring reliable detection at night and in the case of extremely fuzzy visual clues. To solve these problems, we propose a stair detection network with red-green-blue (RGB) and depth inputs. Specifically, we design a selective module, which can make the network learn the complementary relationship between the RGB feature maps and the depth feature maps and fuse the features effectively in different scenes. In addition, we propose several postprocessing algorithms, including a stair line clustering algorithm and a coordinate transformation algorithm, to obtain the stair geometric parameters. Experiments show that our method has better performance than existing the state-of-the-art deep learning method, and the accuracy, recall, and runtime are improved by 5.64%, 7.97%, and 3.81 ms, respectively. The improved indexes show the effectiveness of the multimodal inputs and the selective module. The estimation values of stair geometric parameters have root mean square errors within 15 mm when ascending stairs and 25 mm when descending stairs. Our method also has extremely fast detection speed, which can meet the requirements of most real-time applications.

show abstract

Image Segmentation of Cabin Assembly Scene Based on Improved RGB-D Mask R-CNN

Cited by 13 publications

References 34 publications

Individual Tree Species Identification for Complex Coniferous and Broad-Leaved Mixed Forests Based on Deep Learning Combined with UAV LiDAR Data and RGB Images

Individual Tree Species Identification for Complex Coniferous and Broad-Leaved Mixed Forests Based on Deep Learning Combined with UAV LiDAR Data and RGB Images

Advancing image segmentation with DBO-Otsu: Addressing rubber tree diseases through enhanced threshold techniques

RGB-D-Based Stair Detection and Estimation Using Deep Learning

Contact Info

Product

Resources

About